Search
Close
Search
Advanced Search
Search Menu
AI Discovery Assistant

Abstract

STUDY QUESTION

Is the risk of juvenile idiopathic arthritis (JIA) increased in children conceived after fertility treatment, and is an observed association caused by specific types of fertility treatment or by factors associated with the underlying infertility?

SUMMARY ANSWER

The risk of JIA in children conceived after fertility treatment (any and specific types of fertility treatment) was not convincingly affected when compared with children born to fertile women.

WHAT IS KNOWN ALREADY

It has been suggested that fertility treatment may affect the development of the immune system and thereby increase the risk of developing autoimmune diseases, including JIA.

STUDY DESIGN, SIZE, DURATION

This retrospective population-based cohort study included all live-born children in Denmark between 1 January 1996 and 31 December 2012 (n = 1 084 184). The study population was followed from date of birth until first diagnosis of JIA as registered in the Danish National Patient Registry, date of 16th birthday, date of emigration, date of death or end of follow-up (31 December 2014), whichever occurred first.

PARTICIPANTS/MATERIALS, SETTING, METHODS

The study cohort was linked to the Danish Infertility Cohort in order to identify children born to women with fertility problems (n = 174 702) and fertility treatment (n = 89 931). Cox proportional hazard regression models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) adjusted for potential confounders.

MAIN RESULTS AND THE ROLE OF CHANCE

During a median follow-up period of 10.3 years, 2237 children were diagnosed with JIA. Children born to women with fertility problems had an increased risk of JIA (HR 1.18, 95% CI 1.05–1.32) compared with children born to fertile women. However, the risk was not increased in children conceived after any fertility treatment (HR 1.11; 95% CI 0.95–1.29), or after specific types of fertility treatment being ART (HR 1.05; 95% CI 0.83–1.33), IVF (HR 1.01; 95% CI 0.73–1.38), ICSI (HR 0.98; 95% CI 0.64–1.50) or any fertility drugs (HR 1.10; 95% CI 0.94–1.28) compared with children born to fertile women. The associations between fertility treatment and JIA were also assessed by using children born to women with fertility problems without fertility treatment in the index pregnancy as a reference group, however, the findings did not change substantially.

LIMITATIONS REASONS FOR CAUTION

Despite a large study population, the statistical precision in some subgroup analyses may be affected due to the low number of JIA cases. There may be some misclassification of fertility problems, as some women have undiagnosed fertility problems and are therefore not included in the Danish Infertility Cohort; potentially leading to slight attenuation of the association between fertility problems and JIA.

WIDER IMPLICATIONS OF THE FINDINGS

The results are based on national data and our findings can therefore be applied to other similar populations. Our results indicate that fertility treatment per se do not increase the risk of JIA but merely that the increased risk of JIA observed among children born to women with fertility problems may be due to underlying factors related to both infertility and JIA. However, as this is the first large study in this field, further studies are needed to confirm our findings.

STUDY FUNDING/COMPETING INTEREST(S)

The study was supported by grants from the Jascha Foundation, the Aase and Ejner Danielsens Foundation and The Danish Rheumatism Association. All authors report no conflicts of interest.

TRIAL REGISTRATION NUMBER

N/A.

infertility, fertility treatment, assisted reproduction, in vitro fertilisation, child follow-up, juvenile idiopathic arthritis, cohort study, epidemiology, Denmark

Introduction

Infertility, defined as failure to conceive spontaneously within 12 months of regular unprotected intercourse, affects one in six couples worldwide (Boivin et al., 2007). In Denmark, which has one of the highest proportions of children born after fertility treatment in the Western world, 9% of all children were conceived after fertility treatment in 2016 (Danish Ferility Society).

With the increasing use of fertility treatment follows a public responsibility to continuously monitor potential adverse health outcomes in children born after these treatment procedures, including ovulation induction (OI), IUI and ART counting IVF and ICSI. Several studies have observed an increased risk of adverse perinatal outcomes (e.g. preterm birth, low birth weight and birth defects) in children conceived after fertility treatment compared with children conceived spontaneously (McDonald et al., 2010; Pandey et al., 2012; Hansen et al., 2013), whereas fewer studies have focused on the potential long-term adverse health outcomes including childhood cancer, asthma, cerebral palsy and mental disorders (Klemetti et al., 2010; Bay et al., 2013; Hargreave et al., 2013; Kallen, 2014; Kettner et al., 2015).

The autoimmune disease juvenile idiopathic arthritis (JIA) is defined as all forms of arthritis with onset before the age of 16 and persistent for at least 6 weeks (Prakken et al., 2011). The annual incidence rate of JIA is 15 per 100 000 children in the Nordic countries (Berntson et al., 2003). The disease causes pain and disability and the patients are at risk of an overall sedentary lifestyle and increased morbidity in adulthood (Norgaard et al., 2016). The aetiology is poorly understood but it has been hypothesised that JIA is the result of an abnormal functioning immune system caused by both genetic and environmental components (Ravelli and Martini, 2007). It has been suggested that early life exposures in utero may affect the development of the immune system and thereby increase the likelihood of autoimmune diseases later in life including JIA (Chen et al., 2016). Fertility treatment, including hormonal stimulation with fertility drugs and the chemical and mechanical procedures involved in ART can potentially adversely affect the development of the immune system by inducing epigenetic alterations around the time of conception (Chen et al., 2016) or by bypassing the natural selection of the gametes (Hanevik et al., 2016). Lastly, it is also possible that underlying environmental, biological and genetic factors related both to infertility and JIA may affect the risk of JIA among children conceived to women with fertility problems.

Although an association between infertility, fertility treatment and JIA and other autoimmune diseases can be hypothesised, the number of studies are limited and have focused primarily on type 1 diabetes mellitus (Klemetti et al., 2010; Hargreave et al., 2016; Kettner et al., 2016). Only one Finnish study by Koivurova et al. (2003) has examined the association between fertility treatment and risk of JIA, but the study is limited by a very low number of exposed JIA cases and short follow-up (Koivurova et al., 2003).

We therefore performed a large population-based cohort study including all live-born children in Denmark between 1996 and 2012 to examine whether children conceived after fertility treatment have an increased risk of JIA compared with children born to fertile women, while also taking into account the effects of the underlying fertility problems.

Materials and Methods

The present study is based on several Danish registries and the Danish Infertility Cohort. A detailed description of all data sources is provided in Supplementary Data Files 1 (the Danish registries) and 2 (the Danish Infertility Cohort).

Study population

From the Medical Birth Registry, we identified all 1 091 251 live births in Denmark between 1 January 1996 and 31 December 2012. We excluded children with missing information on sex (n = 59), and maternal age at birth (n = 27). Further, pregnancies of <22 or >44 completed weeks of gestation (n = 6016) and children who died immediately after birth (i.e. on the same date) (n = 965) were excluded, leaving 1 084 184 children for analysis.

Information on fertility problems and fertility treatment

We used the unique personal identification number assigned to all Danish residents to link children in the study cohort to their respective mothers in the Danish Infertility Cohort to identify children conceived by women with fertility problems and children conceived after fertility treatment. The Danish Infertility Cohort includes information on virtually all women in Denmark evaluated for fertility problems in the period 1963–2012 obtained from local computerised systems and medical records, the Danish National Patient Registry, the Danish IVF Registry, and the Danish National Prescription Registry (Supplementary Data File 1). All children born to a woman registered in the Danish Infertility Cohort (prior birth) were defined as children born to women with fertility problems, whereas children born to women not registered in the cohort (prior birth) were defined as children born to fertile women. Furthermore, we used information from the Danish IVF Registry and the Danish Prescription Registry to identify the fertility treatment. We determined the date of conception for all study cohort children by subtracting the gestational length from the child’s date of birth as registered in the Danish Medical Birth Registry and adding 14 days. Children conceived after ART was defined by an ART registration in the Danish IVF Registry 2 weeks before or 2 weeks after the estimated date of conception. We defined children conceived after use of fertility drugs based on a registration of fertility drugs in the Danish IVF Registry and/or a redeemed prescription of fertility drugs in the Danish Prescription Registry. In the Danish IVF registry, fertility drugs used in the index pregnancy was defined as 14 days before the conception date to 16 weeks after, whereas the timeframe was 12 weeks before to 16 weeks after the woman´s estimated conception date in the Danish Prescription Registry.

Outcome

From the Danish National Patient Registry, information on children in the study cohort diagnosed with JIA (ICD-10 code DM08) were retrieved. The study population was followed from date of birth until first diagnosis of JIA, 16th birthday, date of emigration, date of death or end of follow-up (31 December 2014), whichever occurred first.

Potential confounders

Potential confounders were selected a priori based on a literature obtained knowledge of their independent influence on both the exposure and the outcome, albeit limited by their availability in the national Danish registries. From the Danish Medical Birth Registry, we obtained information on year of birth (5-years intervals), maternal and paternal age at birth (quartiles), maternal smoking in pregnancy (yes/no) and birth order (1, 2+). From the Population Education Registry, we retrieved information on highest obtained level of maternal and paternal education (high school or less, short higher education, long higher education). Finally, information on parental history of arthritis (yes/no), including JIA (ICD-8 code 712.0, ICD-10 code M08) and/or other types of arthritis (ICD-8 codes 710–718 [excluding 712.0 and 712.09], ICD-10 codes M09-M19) was extracted from the Danish National Patient Registry. As this registry only holds information from 1977 and onwards, this information was only obtained for a subpopulation where both parents were born after 1972.

Statistical analyses

To estimate the risk of JIA, here represented by hazard rates (HRs), we used Cox proportional hazard regression models to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for the association between exposure groups and JIA with the children’s age used as the underlying time scale. For analyses, the following exposure groups were considered: (1) children born to women with fertility problems, (2) children conceived after any fertility treatment (any ART and/or any fertility drugs) used in index pregnancy, (3) children conceived after any or specific types (IVF, ICSI and other ART [frozen embryo replacement (FER), egg donation (ED), testicular sperm aspiration (TESA) or ART, not otherwise specified]) of ART used in index pregnancy and (4) children conceived after use of any or specific types (clomiphene citrate, gonadotropins [follicle-stimulating hormone (FSH) and human menopausal gonadotrophin (hMG)], human chorionic gonadotropin (hCG), gonadotropin-releasing hormone analogues (GnRH) and progesterone) of fertility drugs used in index pregnancy. The vast majority of women treated with fertility drugs received more than one type of fertility drug per treatment cycle. Hence, the risk estimates associated with each type of fertility drug does not represent the independent effect of the specific fertility drug, but merely the marginal effect, i.e. the effect of the fertility drug in question plus the effect of any other types of fertility drugs received. In the main analyses, the risk of JIA in the different exposure groups were compared with children born to fertile women. In order to examine whether an observed effect could be explained by factors associated with the underlying infertility, we also examined the associations between fertility treatment and JIA using children born to women with fertility problems without fertility treatment in the index pregnancy as a reference group.

Using a complete-case dataset, we assessed risk estimates for children born to women with fertility problems and children conceived after any fertility treatment with and without adjustment for the potential confounders included both separately and jointly. Confounders associated with a change of >10% in the HR estimate were retained in further analyses. Although, none of the potential confounders affected the association between fertility problems, any fertility treatment and JIA, we decided to adjust the analyses for year of delivery and birth order. Adjustment for year of delivery, was performed by permitting separate baselines for each quartile while estimating a common effect of the exposure group of interest. This entails that children are compared to children born within a similar calendar-time period. Birth order was included as an adjustment factor to account for the so-called ‘Hygiene hypothesis’ (Bach, 2002) suggesting that early exposure to infections can decrease the risk of autoimmune diseases (including JIA) later in life due to modulation of the developing immune system. Having older siblings in the household may increase the risk of contracting infections in infancy, and as 63% of the children conceived following fertility treatment in our study were firstborn compared to only 25% of the children born to women with fertility problems without fertility treatment in the index pregnancy, it is possible that a potential association between fertility treatment and JIA may be missed by not adjusting for birth order. We evaluated the proportional hazards assumption by scaled Schoenfeld residuals and the graphical plots indicated no violation of the assumption. We used the statistical environment R, version 3.2.3 (R Core Team) with the survival package for all statistical analyses (Therneau).

Ethical approval

The study was approved by the Danish Health Data Board and Statistics Denmark (J.nr. 704645), the Danish Data Protection Agency (J.nr. 2012-41-0770) and the Local Ethical Committees for Copenhagen and Frederiksberg municipalities (J.nr. 01–298/96).

Results

Characteristics of the study population

The study population comprised 1 084 184 children born alive in Denmark between 1 January 1996 and 31 December 2012. A total of 909 482 (83.9%) were born to fertile women and 174 702 (16.1%) children were born to women with fertility problems. Of the latter group, 89 931 children (51.5%) were conceived after fertility treatment in the index pregnancy and 84 771 (48.5%) children were born without fertility treatment in the index pregnancy. During a median follow-up period of 10.3 years (interquartile range [IQR]: 6.0–14.7 years), 2237 children were diagnosed with JIA at a median age of 6.3 years (IQR: 3.1–10.6 years).

Children born to women with fertility problems were more likely to have parents with a higher age and a higher educational level than children born to fertile women. In contrast, maternal smoking in pregnancy was less prevalent for children born to women with fertility problems (Table I).

Table I
Open in new tab

Characteristics of the 1 084 184 live-born children in Denmark (1996–2012) according to fertility problems and fertility treatment.

Children born to fertile womenChildren born to women with fertility problems
Allwith fertility treatment in index pregnancywithout fertility treatment in index pregnancy
n (%)n (%)n (%)n (%)
Total909 482 (83.9a)174 702 (16.1a)89 931 (8.3 a)84 771 (7.8 a)
Year of delivery
 1996–2000291 447 (32.0)39 740 (22.7)21 369 (23.8)18 371 (21.7)
 2001–2004215 796 (23.7)41 112 (23.5)20 562 (22.9)20 551 (24.2)
 2005–2008208 869 (23.0)46 474 (26.6)23 292 (25.9)23 182 (27.3)
 2009–2012193 370 (21.3)47 376 (27.1)24 708 (27.5)22 668 (26.7)
Birth order
 1398 047 (43.8)77 362 (44.3)56 346 (62.7)21 016 (24.8)
 2+511 435 (56.2)97 340 (55.7)33 585 (37.3)63 756 (75.2)
Maternal smoking in pregnancy
 No671 986 (73.9)139 456 (79.8)73 835 (82.1)65 621 (77.4)
 Yes151 977 (16.7)23 650 (13.5)9190 (10.2)14 460 (17.1)
 Missing85 519 (9.4)11 596 (6.6)6906 (7.7)4690 (5.5)
Maternal educational level
 High school or less287 134 (31.6)41 598 (23.8)18 108 (20.1)23 491 (27.7)
 Short higher education312 697 (34.4)66 497 (38.1)34 159 (38.0)32 338 (38.1)
 Long higher education275 521 (30.3)64 186 (36.7)36 451 (40.5)27 735 (32.7)
 Missing34 130 (3.8)2421 (1.4)1213 (1.3)1208 (1.4)
Paternal educational level
 High school or less251 916 (27.7)40 192 (23.0)18 634 (20.7)21 559 (25.4)
 Short higher education403 805 (44.4)81 624 (46.7)41 586 (46.2)40 038 (47.2)
 Long higher education206 839 (22.7)46 328 (26.5)26 220 (29.2)20 108 (23.7)
 Missing46 922 (5.2)6558 (3.8)3491 (3.9)3067 (3.6)
Maternal age at birth
 Median, years (Interquartile range)29.9 (26.7–33.2)32.6 (29.5–35.7)32.6 (29.7–35.9)32.5 (29.4–35.6)
Paternal age at birth
 Median, years (Interquartile range)32.1 (28.8–35.9)34.6 (31.3–38.3)34.6 (31.5–38.4)34.5 (31.2–38.2)
Children born to fertile womenChildren born to women with fertility problems
Allwith fertility treatment in index pregnancywithout fertility treatment in index pregnancy
n (%)n (%)n (%)n (%)
Total909 482 (83.9a)174 702 (16.1a)89 931 (8.3 a)84 771 (7.8 a)
Year of delivery
 1996–2000291 447 (32.0)39 740 (22.7)21 369 (23.8)18 371 (21.7)
 2001–2004215 796 (23.7)41 112 (23.5)20 562 (22.9)20 551 (24.2)
 2005–2008208 869 (23.0)46 474 (26.6)23 292 (25.9)23 182 (27.3)
 2009–2012193 370 (21.3)47 376 (27.1)24 708 (27.5)22 668 (26.7)
Birth order
 1398 047 (43.8)77 362 (44.3)56 346 (62.7)21 016 (24.8)
 2+511 435 (56.2)97 340 (55.7)33 585 (37.3)63 756 (75.2)
Maternal smoking in pregnancy
 No671 986 (73.9)139 456 (79.8)73 835 (82.1)65 621 (77.4)
 Yes151 977 (16.7)23 650 (13.5)9190 (10.2)14 460 (17.1)
 Missing85 519 (9.4)11 596 (6.6)6906 (7.7)4690 (5.5)
Maternal educational level
 High school or less287 134 (31.6)41 598 (23.8)18 108 (20.1)23 491 (27.7)
 Short higher education312 697 (34.4)66 497 (38.1)34 159 (38.0)32 338 (38.1)
 Long higher education275 521 (30.3)64 186 (36.7)36 451 (40.5)27 735 (32.7)
 Missing34 130 (3.8)2421 (1.4)1213 (1.3)1208 (1.4)
Paternal educational level
 High school or less251 916 (27.7)40 192 (23.0)18 634 (20.7)21 559 (25.4)
 Short higher education403 805 (44.4)81 624 (46.7)41 586 (46.2)40 038 (47.2)
 Long higher education206 839 (22.7)46 328 (26.5)26 220 (29.2)20 108 (23.7)
 Missing46 922 (5.2)6558 (3.8)3491 (3.9)3067 (3.6)
Maternal age at birth
 Median, years (Interquartile range)29.9 (26.7–33.2)32.6 (29.5–35.7)32.6 (29.7–35.9)32.5 (29.4–35.6)
Paternal age at birth
 Median, years (Interquartile range)32.1 (28.8–35.9)34.6 (31.3–38.3)34.6 (31.5–38.4)34.5 (31.2–38.2)

aRow percentages. All other percentages in the table are column percentages.

Table I
Open in new tab

Characteristics of the 1 084 184 live-born children in Denmark (1996–2012) according to fertility problems and fertility treatment.

Children born to fertile womenChildren born to women with fertility problems
Allwith fertility treatment in index pregnancywithout fertility treatment in index pregnancy
n (%)n (%)n (%)n (%)
Total909 482 (83.9a)174 702 (16.1a)89 931 (8.3 a)84 771 (7.8 a)
Year of delivery
 1996–2000291 447 (32.0)39 740 (22.7)21 369 (23.8)18 371 (21.7)
 2001–2004215 796 (23.7)41 112 (23.5)20 562 (22.9)20 551 (24.2)
 2005–2008208 869 (23.0)46 474 (26.6)23 292 (25.9)23 182 (27.3)
 2009–2012193 370 (21.3)47 376 (27.1)24 708 (27.5)22 668 (26.7)
Birth order
 1398 047 (43.8)77 362 (44.3)56 346 (62.7)21 016 (24.8)
 2+511 435 (56.2)97 340 (55.7)33 585 (37.3)63 756 (75.2)
Maternal smoking in pregnancy
 No671 986 (73.9)139 456 (79.8)73 835 (82.1)65 621 (77.4)
 Yes151 977 (16.7)23 650 (13.5)9190 (10.2)14 460 (17.1)
 Missing85 519 (9.4)11 596 (6.6)6906 (7.7)4690 (5.5)
Maternal educational level
 High school or less287 134 (31.6)41 598 (23.8)18 108 (20.1)23 491 (27.7)
 Short higher education312 697 (34.4)66 497 (38.1)34 159 (38.0)32 338 (38.1)
 Long higher education275 521 (30.3)64 186 (36.7)36 451 (40.5)27 735 (32.7)
 Missing34 130 (3.8)2421 (1.4)1213 (1.3)1208 (1.4)
Paternal educational level
 High school or less251 916 (27.7)40 192 (23.0)18 634 (20.7)21 559 (25.4)
 Short higher education403 805 (44.4)81 624 (46.7)41 586 (46.2)40 038 (47.2)
 Long higher education206 839 (22.7)46 328 (26.5)26 220 (29.2)20 108 (23.7)
 Missing46 922 (5.2)6558 (3.8)3491 (3.9)3067 (3.6)
Maternal age at birth
 Median, years (Interquartile range)29.9 (26.7–33.2)32.6 (29.5–35.7)32.6 (29.7–35.9)32.5 (29.4–35.6)
Paternal age at birth
 Median, years (Interquartile range)32.1 (28.8–35.9)34.6 (31.3–38.3)34.6 (31.5–38.4)34.5 (31.2–38.2)
Children born to fertile womenChildren born to women with fertility problems
Allwith fertility treatment in index pregnancywithout fertility treatment in index pregnancy
n (%)n (%)n (%)n (%)
Total909 482 (83.9a)174 702 (16.1a)89 931 (8.3 a)84 771 (7.8 a)
Year of delivery
 1996–2000291 447 (32.0)39 740 (22.7)21 369 (23.8)18 371 (21.7)
 2001–2004215 796 (23.7)41 112 (23.5)20 562 (22.9)20 551 (24.2)
 2005–2008208 869 (23.0)46 474 (26.6)23 292 (25.9)23 182 (27.3)
 2009–2012193 370 (21.3)47 376 (27.1)24 708 (27.5)22 668 (26.7)
Birth order
 1398 047 (43.8)77 362 (44.3)56 346 (62.7)21 016 (24.8)
 2+511 435 (56.2)97 340 (55.7)33 585 (37.3)63 756 (75.2)
Maternal smoking in pregnancy
 No671 986 (73.9)139 456 (79.8)73 835 (82.1)65 621 (77.4)
 Yes151 977 (16.7)23 650 (13.5)9190 (10.2)14 460 (17.1)
 Missing85 519 (9.4)11 596 (6.6)6906 (7.7)4690 (5.5)
Maternal educational level
 High school or less287 134 (31.6)41 598 (23.8)18 108 (20.1)23 491 (27.7)
 Short higher education312 697 (34.4)66 497 (38.1)34 159 (38.0)32 338 (38.1)
 Long higher education275 521 (30.3)64 186 (36.7)36 451 (40.5)27 735 (32.7)
 Missing34 130 (3.8)2421 (1.4)1213 (1.3)1208 (1.4)
Paternal educational level
 High school or less251 916 (27.7)40 192 (23.0)18 634 (20.7)21 559 (25.4)
 Short higher education403 805 (44.4)81 624 (46.7)41 586 (46.2)40 038 (47.2)
 Long higher education206 839 (22.7)46 328 (26.5)26 220 (29.2)20 108 (23.7)
 Missing46 922 (5.2)6558 (3.8)3491 (3.9)3067 (3.6)
Maternal age at birth
 Median, years (Interquartile range)29.9 (26.7–33.2)32.6 (29.5–35.7)32.6 (29.7–35.9)32.5 (29.4–35.6)
Paternal age at birth
 Median, years (Interquartile range)32.1 (28.8–35.9)34.6 (31.3–38.3)34.6 (31.5–38.4)34.5 (31.2–38.2)

aRow percentages. All other percentages in the table are column percentages.

Associations between fertility problems, fertility treatment and JIA

Compared to children born to fertile women, children born to women with fertility problems had an increased risk for JIA with a HR of 1.18 (95% CI 1.05–1.32) (Table II). However, neither any fertility treatment used in index pregnancy (HR 1.11; 95% CI 0.95–1.29), any ART used in index pregnancy (HR 1.05; 95% CI 0.83–1.33) nor any type of fertility drugs used in index pregnancy (HR 1.10; 95% CI 0.94–1.28) was associated with JIA when compared with children born to fertile women (Table II). Further, no association between IVF or ICSI and the development of JIA was observed, whereas children conceived after ‘other ART’ had a of 1.58 (95% CI 0.85–2.94). Concerning the specific types of fertility drugs, use of clomiphene citrate was associated with a HR of 1.26 (95% CI 1.01–1.57), whereas no increased risk was seen for any of the other types of fertility drugs (Table II).

Table II
Open in new tab

Adjusted hazard ratios (HRs) and 95% confidence intervals for juvenile idiopathic arthritis (JIA) according to fertility problems and fertility treatment compared to children born to fertile women.

ChildrenJIAPerson-years at riskAdjusted
nnHRa (95% CI)
Children born to fertile women909 4821 8589 268 2941 (reference)
Children born to women with fertility problems174 7023791 626 3851.18 (1.05–1.32)
 Children conceived after fertility treatment89 931182839 2391.11 (0.95–1.29)
  Children conceived after ART37 13071346 9881.05 (0.83–1.33)
   IVF19 43639197 0261.01 (0.73–1.38)
   ICSI13 40522116 7570.98 (0.64–1.50)
   Other ARTb42891033 2051.58 (0.85–2.94)
 Children conceived after fertility drugsc89 285179831 7801.10 (0.94–1.28)
  Clomiphene citrate33 82482332 0191.26 (1.01–1.57)
  Gonadotropins57 10299509 8931.00 (0.82–1.23)
  hCG68 146124606 3561.05 (0.88–1.27)
  GnRH38 62265350 3960.96 (0.75–1.23)
  Progesterone37 07764356 1250.92 (0.71–1.18)
ChildrenJIAPerson-years at riskAdjusted
nnHRa (95% CI)
Children born to fertile women909 4821 8589 268 2941 (reference)
Children born to women with fertility problems174 7023791 626 3851.18 (1.05–1.32)
 Children conceived after fertility treatment89 931182839 2391.11 (0.95–1.29)
  Children conceived after ART37 13071346 9881.05 (0.83–1.33)
   IVF19 43639197 0261.01 (0.73–1.38)
   ICSI13 40522116 7570.98 (0.64–1.50)
   Other ARTb42891033 2051.58 (0.85–2.94)
 Children conceived after fertility drugsc89 285179831 7801.10 (0.94–1.28)
  Clomiphene citrate33 82482332 0191.26 (1.01–1.57)
  Gonadotropins57 10299509 8931.00 (0.82–1.23)
  hCG68 146124606 3561.05 (0.88–1.27)
  GnRH38 62265350 3960.96 (0.75–1.23)
  Progesterone37 07764356 1250.92 (0.71–1.18)

aAdjusted for year of delivery and birth order.

bOf the 4289 children conceived after ‘other ART’, 3354 (78.2%) children were conceived after use of frozen embryo replacement (FER), 369 (8.6%) children were conceived after egg donation (ED), 454 (10.6%) children were conceived after testicular sperm aspiration (TESA) and 112 (2.6%) children were conceived after ART, not specified.

cfertility drugs: clomiphene, gonadotrophins (human menopausal gonadotrophins and follicle-stimulating hormone) and gonadotropin-releasing hormone (GnRH), human chorionic gonadotropin (hCG), progesterone, oestrogen and other/unspecified drugs.

Table II
Open in new tab

Adjusted hazard ratios (HRs) and 95% confidence intervals for juvenile idiopathic arthritis (JIA) according to fertility problems and fertility treatment compared to children born to fertile women.

ChildrenJIAPerson-years at riskAdjusted
nnHRa (95% CI)
Children born to fertile women909 4821 8589 268 2941 (reference)
Children born to women with fertility problems174 7023791 626 3851.18 (1.05–1.32)
 Children conceived after fertility treatment89 931182839 2391.11 (0.95–1.29)
  Children conceived after ART37 13071346 9881.05 (0.83–1.33)
   IVF19 43639197 0261.01 (0.73–1.38)
   ICSI13 40522116 7570.98 (0.64–1.50)
   Other ARTb42891033 2051.58 (0.85–2.94)
 Children conceived after fertility drugsc89 285179831 7801.10 (0.94–1.28)
  Clomiphene citrate33 82482332 0191.26 (1.01–1.57)
  Gonadotropins57 10299509 8931.00 (0.82–1.23)
  hCG68 146124606 3561.05 (0.88–1.27)
  GnRH38 62265350 3960.96 (0.75–1.23)
  Progesterone37 07764356 1250.92 (0.71–1.18)
ChildrenJIAPerson-years at riskAdjusted
nnHRa (95% CI)
Children born to fertile women909 4821 8589 268 2941 (reference)
Children born to women with fertility problems174 7023791 626 3851.18 (1.05–1.32)
 Children conceived after fertility treatment89 931182839 2391.11 (0.95–1.29)
  Children conceived after ART37 13071346 9881.05 (0.83–1.33)
   IVF19 43639197 0261.01 (0.73–1.38)
   ICSI13 40522116 7570.98 (0.64–1.50)
   Other ARTb42891033 2051.58 (0.85–2.94)
 Children conceived after fertility drugsc89 285179831 7801.10 (0.94–1.28)
  Clomiphene citrate33 82482332 0191.26 (1.01–1.57)
  Gonadotropins57 10299509 8931.00 (0.82–1.23)
  hCG68 146124606 3561.05 (0.88–1.27)
  GnRH38 62265350 3960.96 (0.75–1.23)
  Progesterone37 07764356 1250.92 (0.71–1.18)

aAdjusted for year of delivery and birth order.

bOf the 4289 children conceived after ‘other ART’, 3354 (78.2%) children were conceived after use of frozen embryo replacement (FER), 369 (8.6%) children were conceived after egg donation (ED), 454 (10.6%) children were conceived after testicular sperm aspiration (TESA) and 112 (2.6%) children were conceived after ART, not specified.

cfertility drugs: clomiphene, gonadotrophins (human menopausal gonadotrophins and follicle-stimulating hormone) and gonadotropin-releasing hormone (GnRH), human chorionic gonadotropin (hCG), progesterone, oestrogen and other/unspecified drugs.

The associations between fertility treatment and JIA were also assessed using children born to women with fertility problems, but without fertility treatment in the index pregnancy, as a reference group (Table III). Compared to the HRs observed in Table II with children born to fertile women as the reference group, the HR estimates were generally somewhat lower. Concerning the specific types of fertility drugs, the risk of JIA was decreased among children conceived after gonadotropins (HR 0.80; 95% CI 0.63–1.02), GnRH (HR 0.76; 95% CI 0.57–1.01) and progesterone (HR 0.73; 95% CI 0.55–0.97).

Table III
Open in new tab

Adjusted hazard ratios (HRs) and 95% confidence intervals for juvenile idiopathic arthritis (JIA) according to fertility treatment compared to children born to women with fertility problems without fertility treatment in the index pregnancy

ChildrenJIAPerson-years at riskAdjusted
nnHRa (95% CI)
Children born to women with fertility problems without fertility treatment in the index pregnancy84 771197839 2391 (reference)
 Children conceived after fertility treatment89 931182839 2390.89 (0.72–1.09)
  Children conceived after ART37 13071346 9880.84 (0.64–1.10)
   IVF19 43639197 0260.80 (0.57–1.13)
   ICSI13 40522116 7570.78 (0.50–1.22)
   Other ARTb42891033 2051.26 (0.67–2.38)
 Children conceived after fertility drugsc89 285179831 7800.88 (0.72–1.08)
  Clomiphene citrate33 82482332 0191.00 (0.77–1.30)
  Gonadotropins57 10299509 8930.80 (0.63–1.02)
  hCG68 146124606 3560.84 (0.67–1.06)
  GnRH38 62265350 3960.76 (0.57–1.01)
  Progesterone37 07764356 1250.73 (0.55–0.97)
ChildrenJIAPerson-years at riskAdjusted
nnHRa (95% CI)
Children born to women with fertility problems without fertility treatment in the index pregnancy84 771197839 2391 (reference)
 Children conceived after fertility treatment89 931182839 2390.89 (0.72–1.09)
  Children conceived after ART37 13071346 9880.84 (0.64–1.10)
   IVF19 43639197 0260.80 (0.57–1.13)
   ICSI13 40522116 7570.78 (0.50–1.22)
   Other ARTb42891033 2051.26 (0.67–2.38)
 Children conceived after fertility drugsc89 285179831 7800.88 (0.72–1.08)
  Clomiphene citrate33 82482332 0191.00 (0.77–1.30)
  Gonadotropins57 10299509 8930.80 (0.63–1.02)
  hCG68 146124606 3560.84 (0.67–1.06)
  GnRH38 62265350 3960.76 (0.57–1.01)
  Progesterone37 07764356 1250.73 (0.55–0.97)

aAdjusted for year of delivery and birth order.

bOf the 4289 children conceived after ‘other ART’, 3354 (78.2%) children were conceived after use of frozen embryo replacement (FER), 369 (8.6%) children were conceived after egg donation (ED), 454 (10.6%) children were conceived after testicular sperm aspiration (TESA) and 112 (2.6%) children were conceived after ART, not specified.

cfertility drugs: clomiphene, gonadotrophins (human menopausal gonadotrophins and follicle-stimulating hormone) and gonadotropin-releasing hormone (GnRH), human chorionic gonadotropin (hCG), progesterone, oestrogen and other/unspecified drugs.

Table III
Open in new tab

Adjusted hazard ratios (HRs) and 95% confidence intervals for juvenile idiopathic arthritis (JIA) according to fertility treatment compared to children born to women with fertility problems without fertility treatment in the index pregnancy

ChildrenJIAPerson-years at riskAdjusted
nnHRa (95% CI)
Children born to women with fertility problems without fertility treatment in the index pregnancy84 771197839 2391 (reference)
 Children conceived after fertility treatment89 931182839 2390.89 (0.72–1.09)
  Children conceived after ART37 13071346 9880.84 (0.64–1.10)
   IVF19 43639197 0260.80 (0.57–1.13)
   ICSI13 40522116 7570.78 (0.50–1.22)
   Other ARTb42891033 2051.26 (0.67–2.38)
 Children conceived after fertility drugsc89 285179831 7800.88 (0.72–1.08)
  Clomiphene citrate33 82482332 0191.00 (0.77–1.30)
  Gonadotropins57 10299509 8930.80 (0.63–1.02)
  hCG68 146124606 3560.84 (0.67–1.06)
  GnRH38 62265350 3960.76 (0.57–1.01)
  Progesterone37 07764356 1250.73 (0.55–0.97)
ChildrenJIAPerson-years at riskAdjusted
nnHRa (95% CI)
Children born to women with fertility problems without fertility treatment in the index pregnancy84 771197839 2391 (reference)
 Children conceived after fertility treatment89 931182839 2390.89 (0.72–1.09)
  Children conceived after ART37 13071346 9880.84 (0.64–1.10)
   IVF19 43639197 0260.80 (0.57–1.13)
   ICSI13 40522116 7570.78 (0.50–1.22)
   Other ARTb42891033 2051.26 (0.67–2.38)
 Children conceived after fertility drugsc89 285179831 7800.88 (0.72–1.08)
  Clomiphene citrate33 82482332 0191.00 (0.77–1.30)
  Gonadotropins57 10299509 8930.80 (0.63–1.02)
  hCG68 146124606 3560.84 (0.67–1.06)
  GnRH38 62265350 3960.76 (0.57–1.01)
  Progesterone37 07764356 1250.73 (0.55–0.97)

aAdjusted for year of delivery and birth order.

bOf the 4289 children conceived after ‘other ART’, 3354 (78.2%) children were conceived after use of frozen embryo replacement (FER), 369 (8.6%) children were conceived after egg donation (ED), 454 (10.6%) children were conceived after testicular sperm aspiration (TESA) and 112 (2.6%) children were conceived after ART, not specified.

cfertility drugs: clomiphene, gonadotrophins (human menopausal gonadotrophins and follicle-stimulating hormone) and gonadotropin-releasing hormone (GnRH), human chorionic gonadotropin (hCG), progesterone, oestrogen and other/unspecified drugs.

Discussion

In this large retrospective population-based cohort study including all children born alive in Denmark between 1996 and 2012, we found that children born to women with fertility problems had a slightly increased risk of JIA compared to children born to fertile women. However, neither any fertility treatment nor any of the specific types of fertility treatment including different ARTs and fertility drugs were convincingly associated with JIA in the affected children.

To our knowledge, the association between fertility treatment and JIA has only been investigated in one small population-based cohort study from Finland (Koivurova et al., 2003). Caution should be exerted in the interpretation of the substantial, however not statistically significantly odds ratio of 5.5 (95% CI 0.6–49.9) as it was based on only four cases of exposed children with JIA and had a limited follow-up period (maximum follow-up: 3 years of age). In the present large, nationwide population-based study, we find that children born to women with fertility problems have a slightly increased risk of JIA. This increased risk could seemingly not be explained by use as fertility treatment as our detailed analyses showed that children conceived after any fertility treatment or after specific types of fertility treatment including IVF, ICSI and fertility drugs were not at an increased risk compared to children born to fertile women. The findings did not change substantially when children born to women with fertility problems, but without fertility treatment in the index pregnancy were used as a reference group. Our results thereby imply that fertility treatment per se is not a causal factor for JIA and that the observed increased risk of JIA for children born to infertile women may merely be related to factors related to both fertility problems and JIA. The aetiology of JIA is not well established, but it can be hypothesised that a number of environmental and biological factors such as maternal smoking, maternal BMI, air pollution and infections might be common causes for both fertility problems and JIA (Ellis et al., 2010; Sharp et al., 2017; Franca et al., 2018). Furthermore, it is possible that genetic components both associated with fertility problems and JIA could be involved (Brubaker et al., 2018).

An important strength of the present study is the large number of included study subjects and long follow-up period resulting in high statistical precision in most risk estimates, although the precision in some subgroup analyses may be limited due to the low number of JIA cases. Further strengths of this study include the population-based design, which imply that our results are generalisable to other similar populations, virtually absence of loss to follow-up because of the precise linkage between our cohort and the Danish population-based registries and moreover, the registry-based setup implies no selection or recall bias. In addition, the detailed information on type of fertility treatment allowed us to perform analyses of their potential differential effect on JIA risk. The information on fertility problems and fertility treatment are expected to be of high-quality and coverage as it is based on data from the Danish Infertility Cohort that contains information from a number of high-quality data sources including the Danish National Patient Registry, the Danish IVF Registry and the Danish National Prescription Registry. Since 1994 it has been mandatory to report all IVF treatments in Denmark to the Danish IVF Registry (Blenstrup and Knudsen, 2011). Further, the Danish National Prescription Registry holds records for all drug prescriptions dispensed since 1995 (Pottegard et al., 2017). In addition to the reference group of children born to fertile women, we also used a reference group consisting of children born to women with fertility problems, but without fertility treatment in the index pregnancy, thereby allowing us to minimise the effects of the underlying infertility. Lastly, we were able to evaluate several potential confounders obtained from the Danish population-based health registries.

Some limitations from this study are also noteworthy. First, even though information on fertility problems were obtained from a variety of high-quality nationwide data sources, there may be some misclassification of fertility problems. This is due to the fact some women in the general population of Denmark will have undiagnosed fertility problems. This misclassification of exposure may have led to a slight attenuation of the obtained risk estimates for fertility problems. Second, the severity (e.g. length of unwanted childlessness) of fertility problems may have a differential effect on JIA risk but this information is unfortunately not available in the Danish registries. Third, we had no information on the women’s compliance of redeemed fertility drugs, which could have led to non-differential misclassification, potentially resulting in underestimated risk estimates. However, we would expect that women requesting fertility treatment are highly compliant in order to obtain a much-wanted pregnancy, and thus the impact on our results would likely be small. Fourth, even though many potential confounders were assessed, unmeasured or residual confounding due to for example severity of fertility problems, air pollution and infections cannot be ruled out. Lastly, despite this study is the largest to date, the low number of outcomes in some of the subgroups limited the statistical precision of these risk estimates.

In conclusion, our findings indicate that children born to women with fertility problems may have a slightly increased risk of JIA. In addition, our results indicate that this increased risk is likely not due to the fertility treatment per se but may be explained by underlying environmental, biological and genetic factors related both to infertility in the mother and JIA in the child. We encourage a re-evaluation of the obtained results in further large, cohort studies.

Acknowledgements

The authors thank Peter Bo Aarslev and Minna Ghasemi for help with data management, Merete Kjær Hansen for the statistical data analyses and Anne Katrine Duun-Henriksen for input to the analysis strategy.

Authors’ roles

S.K.K., M.H. and A.J. contributed to the conception and design of the study. All authors analysed and interpreted the data. C.D.S was responsible for writing and editing the manuscript. The manuscript was critically revised by S.K.K., M.H. and A.J. All authors approved the final version of the manuscript.

Funding

The study was supported by a grant from the Jascha Foundation, the Aase and Ejner Danielsens Foundation and The Danish Rheumatism Association (R151-A4610).

Conflict of interest

The authors report no conflict of interest.

References

Bach
JF
.
The effect of infections on susceptibility to autoimmune and allergic diseases
.
N Engl J Med
2002
;
347
:
911
–
920
.

Google Scholar

Crossref
Search ADS
PubMed

 

Bay
B
,
Mortensen
EL
,
Hvidtjorn
D
,
Kesmodel
US
.
Fertility treatment and risk of childhood and adolescent mental disorders: register based cohort study
.
BMJ
2013
;
347
:
f3978
.

Google Scholar

Crossref
Search ADS
PubMed

 

Berntson
L
,
Andersson
BG
,
Fasth
A
,
Herlin
T
,
Kristinsson
J
,
Lahdenne
P
,
Marhaug
G
,
Nielsen
S
,
Pelkonen
P
,
Rygg
M
.
Incidence of juvenile idiopathic arthritis in the Nordic countries. A population based study with special reference to the validity of the ILAR and EULAR criteria
.
J Rheumatol
2003
;
30
:
2275
–
2282
.

Google Scholar

PubMed
OpenURL Placeholder Text

 

Blenstrup
LT
,
Knudsen
LB
.
Danish registers on aspects of reproduction
.
Scand J Public Health
2011
;
39
:
79
–
82
.

Google Scholar

Crossref
Search ADS
PubMed

 

Boivin
J
,
Bunting
L
,
Collins
JA
,
Nygren
KG
.
International estimates of infertility prevalence and treatment-seeking: potential need and demand for infertility medical care
.
Hum Reprod
2007
;
22
:
1506
–
1512
.

Google Scholar

Crossref
Search ADS
PubMed

 

Brubaker
WD
,
Li
S
,
Baker
CL
,
Eisenberg
ML
.
Increased risk of autoimmune disorders in infertile men: analysis of US claims data
.
Andrology
2018
;
6
:
94
–
98
.

Google Scholar

Crossref
Search ADS
PubMed

 

Chen
T
,
Liu
HX
,
Yan
HY
,
Wu
DM
,
Ping
J
.
Developmental origins of inflammatory and immune diseases
.
Mol Hum Reprod
2016
;
22
:
858
–
865
.

Google Scholar

Crossref
Search ADS
PubMed

 

Danish Fertility Society. Results from

2016
. [Availible from: https://fertilitetsselskab.dk/wp-content/uploads/2018/10/dfs2016ny.pdf] (Accessed December 2018).

Ellis
JA
,
Munro
JE
,
Ponsonby
AL
.
Possible environmental determinants of juvenile idiopathic arthritis
.
Rheumatology (Oxford)
2010
;
49
:
411
–
425
.

Google Scholar

Crossref
Search ADS
PubMed

 

Franca
CMP
,
Sallum
AME
,
Braga
ALF
,
Strufaldi
FL
,
Silva
CAA
,
Farhat
SCL
.
Risk factors associated with juvenile idiopathic arthritis: exposure to cigarette smoke and air pollution from pregnancy to disease diagnosis
.
J Rheumatol
2018
;
45
:
248
–
256
.

Google Scholar

Crossref
Search ADS
PubMed

 

Hanevik
HI
,
Hessen
DO
,
Sunde
A
,
Breivik
J
.
Can IVF influence human evolution?
Hum Reprod
2016
;
31
:
1397
–
1402
.

Google Scholar

Crossref
Search ADS
PubMed

 

Hansen
M
,
Kurinczuk
JJ
,
Milne
E
,
de Klerk
N
,
Bower
C
.
Assisted reproductive technology and birth defects: a systematic review and meta-analysis
.
Hum Reprod Update
2013
;
19
:
330
–
353
.

Google Scholar

Crossref
Search ADS
PubMed

 

Hargreave
M
,
Jensen
A
,
Deltour
I
,
Brinton
LA
,
Andersen
KK
,
Kjaer
SK
.
Increased risk for cancer among offspring of women with fertility problems
.
Int J Cancer
2013
;
133
:
1180
–
1186
.

Google Scholar

Crossref
Search ADS
PubMed

 

Hargreave
M
,
Kjaer
SK
,
Jorgensen
ME
,
Jensen
A
.
Type 1 diabetes risk in children born to women with fertility problems: a cohort study in 1.5 million Danish children
.
Acta Obstet Gynecol Scand
2016
;
95
:
1441
–
1446
.

Google Scholar

Crossref
Search ADS
PubMed

 

Kallen
B
.
The risk of neurodisability and other long-term outcomes for infants born following ART
.
Semin Fetal Neonatal Med
2014
;
19
:
239
–
244
.

Google Scholar

Crossref
Search ADS
PubMed

 

Kettner
LO
,
Henriksen
TB
,
Bay
B
,
Ramlau-Hansen
CH
,
Kesmodel
US
.
Assisted reproductive technology and somatic morbidity in childhood: a systematic review
.
Fertil Steril
2015
;
103
:
707
–
719
.

Google Scholar

Crossref
Search ADS
PubMed

 

Kettner
LO
,
Matthiesen
NB
,
Ramlau-Hansen
CH
,
Kesmodel
US
,
Bay
B
,
Henriksen
TB
.
Fertility treatment and childhood type 1 diabetes mellitus: a nationwide cohort study of 565,116 live births
.
Fertil Steril
2016
;
106
:
1751
–
1756
.

Google Scholar

Crossref
Search ADS
PubMed

 

Klemetti
R
,
Sevon
T
,
Gissler
M
,
Hemminki
E
.
Health of children born after ovulation induction
.
Fertil Steril
2010
;
93
:
1157
–
1168
.

Google Scholar

Crossref
Search ADS
PubMed

 

Koivurova
S
,
Hartikainen
AL
,
Sovio
U
,
Gissler
M
,
Hemminki
E
,
Jarvelin
MR
.
Growth, psychomotor development and morbidity up to 3 years of age in children born after IVF
.
Hum Reprod
2003
;
18
:
2328
–
2336
.

Google Scholar

Crossref
Search ADS
PubMed

 

McDonald
SD
,
Han
Z
,
Mulla
S
,
Ohlsson
A
,
Beyene
J
,
Murphy
KE
.
Preterm birth and low birth weight among in vitro fertilization twins: a systematic review and meta-analyses
.
Eur J Obstet Gynecol Reprod Biol
2010
;
148
:
105
–
113
.

Google Scholar

Crossref
Search ADS
PubMed

 

Norgaard
M
,
Twilt
M
,
Andersen
LB
,
Herlin
T
.
Accelerometry-based monitoring of daily physical activity in children with juvenile idiopathic arthritis
.
Scand J Rheumatol
2016
;
45
:
179
–
187
.

Google Scholar

Crossref
Search ADS
PubMed

 

Pandey
S
,
Shetty
A
,
Hamilton
M
,
Bhattacharya
S
,
Maheshwari
A
.
Obstetric and perinatal outcomes in singleton pregnancies resulting from IVF/ICSI: a systematic review and meta-analysis
.
Hum Reprod Update
2012
;
18
:
485
–
503
.

Google Scholar

Crossref
Search ADS
PubMed

 

Pottegard
A
,
Schmidt
SA
,
Wallach-Kildemoes
H
,
Sorensen
HT
,
Hallas
J
,
Schmidt
M
.
Data resource profile: the Danish National Prescription Registry
.
Int J Epidemiol
2017
;
46
:
798f
.

Google Scholar

OpenURL Placeholder Text

 

Prakken
B
,
Albani
S
,
Martini
A
.
Juvenile idiopathic arthritis
.
Lancet
2011
;
377
:
2138
–
2149
.

Google Scholar

Crossref
Search ADS
PubMed

 

R Core Team
. R: A language and enviroment for statistical computing. R Foundation for for statistical Computing, Vienna, Austria [Available from: http://www.R-project.org/]

Ravelli
A
,
Martini
A
.
Juvenile idiopathic arthritis
.
Lancet
2007
;
369
:
767
–
778
.

Google Scholar

Crossref
Search ADS
PubMed

 

Sharp
GC
,
Salas
LA
,
Monnereau
C
,
Allard
C
,
Yousefi
P
,
Everson
TM
,
Bohlin
J
,
Xu
Z
,
Huang
RC
,
Reese
SE
et al.
Maternal BMI at the start of pregnancy and offspring epigenome-wide DNA methylation: findings from the pregnancy and childhood epigenetics (PACE) consortium
.
Hum Mol Genet
2017
;
26
:
4067
–
4085
.

Google Scholar

Crossref
Search ADS
PubMed

 

Therneau
T.
A Package for Survival Analysis in S, version 2.38. [Available from: https://cran.r-roject.org/web/packages/survival/index.html.], (accessed June 2017).

© The Author(s) 2019. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved. For Permissions, please e-mail: [email protected]
This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
Topic:
Issue Section:
Original Article > Reproductive epidemiology
Download all slides
  • Supplementary data

    • Supplementary data

    SUPPLEMENTARY_DATA_dez035 - pdf file