Associations between maternal complications during pregnancy and childhood asthma: a retrospective cohort study in southern China

Background The associations between maternal complications during pregnancy and childhood asthma have rarely been investigated in low and middle-income countries. We aimed to investigate the associations among the southern Chinese population, and to determine whether the associations were mediated through preterm birth, cesarean delivery, low birth weight and non-breasting feeding in the first 6 months or modified by childs lifestyles (smoking, body mass index(BMI), and sleep duration). Methods We conducted a retrospective cohort study of 208,190 children in Guangzhou, China. Information on maternal gestational hypertension, gestational diabetes, gestational anemia, and hepatitis B during pregnancy was extracted from medical records. Ever diagnosis of asthma of the children was obtained by questionnaire. We used binomial logistic regression models to estimate the adjusted odds ratios (aORs) and 95% confidential interval (CI) for childhood asthma. We conducted mediation analyses to estimate the indirect effects of gestational complications on childhood asthma mediated through preterm birth, cesarean delivery, low birth weight, and non-breastfeeding in the first 6 months. Findings The overall prevalence of ever diagnosed asthma was 1.3%. Gestational hypertension, gestational diabetes, gestational anemia, or hepatitis B during pregnancy was positively associated with ever diagnosed childhood asthma, with the aOR of 1.66(95%CI 1.31-2.10), 1.68(95%CI 1.40-2.02), 1.69(95%CI 1.49-1.91) and 1.54(95%CI 1.13-2.08), respectively. A stronger association was observed for 2 or more gestational complications (aOR= 2.34, 95%CI 1.71-3.23) than 1 gestational complication (aOR=1.80, 95%CI 1.62-1.99). The associations between maternal complications and childhood asthma did not differ by childs smoking, BMI, or sleep duration, except that the aOR for maternal gestational hypertension was significantly higher in children who were active smoker (aOR=5.68, 95%CI 2.09-15.42) than those who were non-smoker (aOR=1.56, 95%CI 1.21-2.00). A small proportion of the associations were mediated through preterm birth, cesarean delivery, low birth weight, and non-breastfeeding in the first 6 months. Interpretation Gestational hypertension, diabetes, anemia, or hepatitis B during pregnancy was significantly associated with childhood asthma in the southern Chinese population, and the associations were partially explained by the mediation effects of cesarean delivery, preterm birth, low birthweight and non-breastfeeding in the first 6 months.

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted March 22, 2022. ; https://doi.org/10.1101/2022.03.21.22272680 doi: medRxiv preprint 4 / 28 67 observed for 2 or more gestational complications (aOR= 2.34, 95%CI 1.71-3.23) than 1 gestational 68 complication (aOR=1.80, 95%CI 1.62-1.99). The associations between maternal complications and 69 childhood asthma did not differ by child's smoking, BMI, or sleep duration, except that the aOR for 70 maternal gestational hypertension was significantly higher in children who were active smoker 71 (aOR=5.68, 95%CI 2.09-15.42) than those who were non-smoker (aOR=1.56, 95%CI 1.21-2.00). A 72 small proportion of the associations were mediated through preterm birth, cesarean delivery, low 73 birth weight, and non-breastfeeding in the first 6 months.  is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted March 22, 2022. ; https://doi.org/10.1101/2022.03.21.22272680 doi: medRxiv preprint

112
Childhood asthma is a public health concern worldwide, with a prevalence of 1.8-27.4% in children 113 aged 6-7 years and 2.6-30.5% in children aged 13-14 years, varying across countries and between 114 sex[1]. Although mortality of childhood asthma is relatively low, childhood asthma represents a 115 significant burden on both the family and society and ranks among the top 20 conditions worldwide 116 for disability-adjusted life years in children [2]. Therefore, it is crucially important to understand the 117 risk factors of childhood asthma. 118 Early stage of life is a sensitive period for the development of respiratory health [3], and several 119 studies examined the associations between maternal complications during pregnancy and asthma risk 120 in the offspring [4][5][6][7][8][9][10][11][12][13][14]. However, findings from these studies were mixed: some studies suggested 121 exposure to maternal preeclampsia was associated with an increased risk of asthma in the 122 offspring[4, 6], while others suggested a null association [5,7,9,14]. Additionally, current 123 knowledge of the associations between maternal complications and childhood asthma is 124 predominantly derived from high-income countries [4][5][6][7][8][9][10][11][12], which may not be generalizable to middle 125 and low-income countries with different epidemiological characteristics of gestational complications. 126 For instance, pregnant women in China have a higher prevalence of maternal anemia and hepatitis B 127 than in high-income countries [15,16]. Moreover, the effects of maternal complications during the 128 pregnancy on childhood asthma may be mediated through birth outcomes and feeding models [5] or 129 modified by child's lifestyles [17][18][19], which have rarely been investigated so far.

130
In light of these considerations, we carried out a retrospective cohort study in southern China to 131 investigate the associations of gestational hypertension, gestational diabetes, gestational anemia, and 132 hepatitis B during pregnancy with asthma among children aged 6 to 14 years, and whether the . CC-BY 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint  . Project-related design, organization, and implementation have been 147 described previously [22]. Parents whose child planned to participate in the annual physical 148 examination were contacted through the internet to get parental consent to grant permission to their 149 children to participate in the PHMPPHSS. A pre-developed health management software was shared 150 with the parents after they provided the written informed consent. The parents then registered in the 151 health management software, in which a structured parental questionnaire and a child's questionnaire 152 were designed to collect data.

153
The parental questionnaire included the profile of the index child (age, sex, number of siblings, 154 residential region, medical history, including asthma diagnosis); mothers' age, educational level, and . CC-BY 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint 155 smoking during pregnancy, family monthly income; and feeding models of the index child in the first 156 6 months. Parental questionnaire also included pre-pregnancy body weight and height of mother; 157 delivery patterns; gestational weeks; complications during pregnancy, and the birth weight of the 158 index child, all of which were asked to answer based on medical records during pregnancy and the 159 index child's birth certificate.   169 This study is reported as per the Strengthening the Reporting of Observational Studies in 170 Epidemiology (STROBE) guideline (S1 Checklist). We conducted a retrospective cohort study in Guangzhou. We excluded 9,064 children whose mothers did not answer the parental questionnaire; 174 63 children whose mothers had pre-existing hypertension or type 1 or 2 diabetes; 6,063 children 175 whose mothers did not respond to gestational complications; 8,678 children whose mothers did not 176 respond to asthma diagnosis of the index child. We further excluded 1,201 children who were . CC-BY 4.0 International license It is made available under a perpetuity.

Study population
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint    is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

219
Binomial logistic regression models were employed to estimate the adjusted odds ratios (aORs) 220 and 95% confidential interval (CI) for childhood asthma by gestational hypertension, gestational . CC-BY 4.0 International license It is made available under a perpetuity.
is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted March 22, 2022. ; https://doi.org/10.1101/2022.03.21.22272680 doi: medRxiv preprint 11 / 28 221 diabetes, gestational anemia, and hepatitis B, respectively, and two stages of adjustment were used to 222 examine the robustness of the associations. We selected potential confounding factors which are 223 known to be associated with maternal complications and the outcomes. In model 1, we adjusted for 224 children's sex (boy or girl), age (continuous variable), residential region (urban or rural), one-child 225 family (yes or no), maternal age (younger than 20 years; 21-35 years; or older than 35 years), 226 mother's educational level (junior high school or below; senior high school; or college or above) and  is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint age and residential region, we also examined whether the association between maternal 245 complications during pregnancy and childhood asthma varied with child's age (6~12 years, or 13~14 246 years) and residential region (urban and rural). Among all the subgroup analyses, we adjusted for the 247 covariates as we did in model 2. Differences within subgroup analyses were assessed by the test for   is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint  Table S1. The age of children ranged from 6 to 14 years, with a mean age of 9.5 (SD=2.5) years. 268 Overall, 2,696 (1.3%, 95%CI 1.2%-1.4%) children had ever been diagnosed with asthma. Compared 269 to children without asthma, children with asthma were more likely to live in urban areas, be boys, 270 older, born through cesarean delivery, preterm birth, and overweight/obesity, and have inadequate 271 sleep duration (Table 1).  Table   279 2. The aORs reduced substantially from model 1 to model 2 except for gestational anemia. In the 280 fully adjusted model (model 2 in  is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted March 22, 2022. ; https://doi.org/10.1101/2022.03.21.22272680 doi: medRxiv preprint 14 / 28 287 exposed to 1, and 2 or more gestational complication (s) was 2.3% and 3.1%, respectively; while the 288 prevalence of asthma among children who did not expose to any gestational complication was 1.2%.

289
The aORs are presented in Table 3. Both the aORs did not change substantially from model 1 to 290 model 3. In the fully adjusted model (model 2 in Table 3), the aOR for 1 gestational complication 291 was 1 .80 (95%CI, 1.62-1.99), and an increased aOR was observed for 2 or more gestational 292 complications (2.34, 95%CI, 1.71-3.23). were also similar according to child's age group, and residential region (See in supplementary Table   300 S2 and S3). is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint In this study among the southern Chinese population, we found that maternal gestational 313 hypertension, gestational diabetes, gestational anemia, or hepatitis B was associated with childhood 314 asthma, respectively. A stronger association was observed for 2 or more gestational complications 315 than 1 gestational complication. The associations were not modified by children's smoking, BMI, 316 sleep duration and child's age group, except that the aOR for maternal gestational hypertension was 317 significantly higher in children who were active smoker than those who were non-smoker. A small 318 proportion of the associations were mediated through cesarean delivery, preterm birth, low 319 birthweight and non-breastfeeding in the first 6 months. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint Four studies examined the association between gestational anemia and childhood asthma, and 336 the results were also mixed[9, 11, 12, 14]. Nafstad et al. [9] found that exposure to gestational 337 anemia was associated with an increased risk of childhood asthma (OR=1.67, 95%CI 1.24-2.36). is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint  is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint  In the southern Chinese population, we found that gestational hypertension, gestational diabetes, 395 gestational anemia, or hepatitis B are significantly associated with childhood asthma, and a stronger 396 association was observed for 2 or more gestational complications than 1 gestational complication. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The associations between maternal gestational complications and childhood asthma were not 398 modified by child's lifestyles, but could be partially mediated through cesarean delivery, preterm 399 birth, low birthweight and non-breastfeeding in the first 6 months.   is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint   is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint    is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint  is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint