Epidemiology of birth defects based on surveillance data from 2011–2015 in Guangxi, China: comparison across five major ethnic groups

Background The causes of birth defects (BDs) are complex and include genetic and environmental factors and/or their interactions. More research is needed to describe the epidemiology of BDs within specific regions of China. This study focused on differences in the prevalence of BDs based on ethnicity in a large city in Guangxi Province, China. Methods Surveillance data of infants born in 114 registered hospitals in Liuzhou between 2011 and 2015 were analyzed to determine the epidemiology of BDs across five major ethnic groups. We calculated the prevalence of BDs and relative risk of BDs by ethnicity. Results There were 260,722 perinatal infants of which 6581 had BDs, with the average prevalence of 25.24 per 1000 perinatal infants (PIs). Prevalence data showed an obvious uptrend over the past 5 years. Han had the highest prevalence of total BDs (28.98‰), followed by Zhuang (25.19‰), Yao (18.50‰), Miao (15.78‰) and Dong (14.24‰). Relative to the Han; Zhuang, Miao, Yao, and Dong had a lower risk of musculoskeletal and urogenital malformations; Miao and Yao had a lower risk of cardiovascular malformation; and Dong had a lower risk of cardiovascular and craniofacial malformation. Several maternal risk factors were found to be associated with BDs (e.g., maternal and gestational age, number of antenatal care visits). Conclusion This study provided a comprehensive description of ethnic differences in the risk of BDs in Liuzhou City, China. Observed ethnic differences in the risk of BDs may be related to genetic susceptibilities, environment, cultural customs, or to potential combinations of these factors.

The causes of BDs are complex and multifactorial. Nearly 50% of BDs cannot be ascribed to a specific cause [8,9]. In 1964, after the thalidomide tragedy, several countries including the Great Britain, Israel, and Finland started conducting BD surveillance [10][11][12][13][14]. The Chinese Ministry of Health started the BD surveillance systems in 1986 [15][16][17]. The surveillance was initially paper-based data reporting method, that was replaced by an electronic, web-based reporting system developed by the National Office for Maternal and Child Health Surveillance in 1998 [18]. The quality of birth defect monitoring varies in different parts of the world, and may vary even within the same country or region [19,20]. In China, the emphasis placed on monitoring for BDs is higher and more comprehensive in Eastern China than in Midwestern China [3].
Guangxi is one of the western provinces of China, and the Chinese minorities account for majority of the population in Guangxi province (37.2%) [21]. Liuzhou is the most representative city in Guangxi, for its geographic (population of 3.8 million), economic (region's industrial center), composition of population (large ethnic minority groups), and culture and customs within the ethnicities [22]. There are fifty-six (56) ethnic groups in China, and more than thirty ethnic groups reside in Liuzhou. The largest ethnic group in Liuzhou is Han (48.9%), followed by Zhuang (35.2%), Miao (6.4%), Dong (6.3%), Yao (1.9%), and Mulao (0.8%) [21,23,24]. The Chinese ethnic groups have their own, different languages, culture, customs and living environments. The epidemiology of BDs among various Chinese ethnic groups has not been described in the literature. Since the etiology of BDs could be multi-factorial involving genetic factors, environmental factors and their interactions, it is crucial to understand the difference in BD phenotypes among Chinese ethnic groups. Therefore, our study aims to provide further insights on BDs among the different Chinese ethic groups in Liuzhou. The information gained may allow clinicians to provide proper counselling to families, allow public health officials to appropriately plan targeted interventions, and may provide the foundation for further etiologic and epidemiologic studies.

Disease classification
The diagnosis of BD was based on the "International Statistical Classification of Diseases and Related Health Problems, Tenth Edition" (ICD-10) and Chinese National Criteria of BDs [25]. BDs were categorized by the system affected, including neurologic system, craniofacial system, gastrointestinal system, urogenital system, musculoskeletal system, cardiovascular system, respiratory system, genetic metabolic diseases, genetic syndrome, and other BDs (i.e., those BDs do not belong to the above-mentioned 9 systems). BDs were classified into ten systems (including neurologic, craniofacial, gastrointestinal, urogenital, musculoskeletal, cardiovascular, respiratory, genetic metabolic, genetic syndrome, and other) in our study. When a patient had BDs affecting ≥2 organ-systems, then he/she was categorized as having multiple BDs.

Data collection
All surveillance data of BDs were collected from the obstetrics departments or neonatal departments according to the "National Office for Maternal and Child Health Surveillance" formulated by the National Health and Family Planning Commission. The surveying population was perinatal infants (including stillbirth, fetal death or live birth between 20 weeks of gestation through 7 days after birth) born in the 114 registered hospitals of Liuzhou from 2011 to 2015. In total, infants were followed six times in the first year of life. The schedule for follow-up monitoring was as follows: Within 7 days for the first time-point, and then at 6 weeks, 3 months, 6 months, 8 months and 12 months old. Infant BDs were reported to the monitoring system when diagnosed within 7 days of life.
Each delivery that was associated with a BD was reported using a registration card for BDs submitted by physicians in obstetrics and gynecology, pediatric or neonatal medicine through an online hospital-based survey, required by the Chinese government. Each case report card recorded basic maternal information, (including ethnicity, residence, family income, education, mother's age, number of antenatal care visits, gestational age at first antenatal care visit, number of reported abortions, pregnancy outcomes, Human Immunodeficiency Virus (HIV), Hepatitis B Virus (HBV), and syphilis infections), birth information (e.g., gestational age, weight of birth), diagnoses of specific BDs, symptoms, medication use during early pregnancy, and family history. In addition to the case report card, a quarterly table for each registered hospital was completed by professional physicians. Each quarterly table contained 3 months of data such as the number of perinatal births, maternal age, residence, ethnicity, occupation, pregnancy history, gestational age of birth, gestational age at first antenatal care attendance, number of antenatal visits, infant gender, number of BDs, and maternal illness.
Both case report cards and quarterly tables were reviewed and audited by maternal and child health hospitals and health administrative departments. Periodic quality control measures were in place at the monitored hospitals and occurred quarterly at the county-level and bi-annually at the city-level or province-level to assure reporting accuracy.

Statistical analysis
The prevalence rates and 95% confidence intervals of overall BDs were calculated for the whole sample and stratified by the five major ethnic groups across all years (2011-2015) of the study. A line chart was then constructed to graphically display the longitudinal trends of the prevalence of BDs by the different ethnic groups. The prevalence rates of the top fifteen BDs were also calculated and compared among the six major ethnic groups using chi-square tests. Multiple logistic regression analysis was performed to further examine the association of ethnicity with BDs, controlling for potential confounders including maternal age, gestational age, number of antenatal care visits, gestational age at first antenatal care visit, number of previous abortion, and syphilis infection. The dependent variable of logistic regression analysis was whether the infant had a birth defect (no/yes). Only variables identified as significantly associated with the dependent variable at P < 0.05 in the bivariate analyses were included in the model. Adjusted odds ratios (ORs) and their 95% confidence intervals (CIs) were calculated. All statistical analyses were performed using the SAS 9.4 statistical software package (SAS Institute Inc., Cary, NC, USA). A significance level of 0.01 was adopted in bivariate comparisons and multivariate analysis due to the large sample size.

Study sample
A total of 260,722 perinatal infants were monitored from 2011 to 2015. Of the perinatal infants, the top 5 ethnic groups were represented as 44.7% were Han, 36.6% were Zhuang, 7.5% were Miao, 6.9% were Dong, and 2.4% were Yao (see Table 1). The rest, 1.8%, included other various, small ethnic groups which are not further discussed.

Prevalence and trend of total BDs
Of the 260,722 perinatal infants, 6581 had BDs, with an average prevalence of 25.2 (95% CI: 24.6-25.9) per 1000 PIs (perinatal infants) ( Table 2). Trend analysis revealed that the annual prevalence rates of total BDs in the 5 years increased linearly (χ 2 trend = 59.92, P < 0.0001). The total prevalence rate increased by 1.49% annually from 2011 to 2015 on average.
From 2011 to 2015, 89 classes (or combinations of classes) of BDs in total were found in our sample, and the top fifteen classes of BDs were musculoskeletal malformation, cardiovascular, craniofacial, urogenital system, other BDs, neurologic, gastrointestinal system, genetic syndrome, respiratory system, musculoskeletal with cardiovascular systems, cardiovascular with other BDs, cardiovascular with urogenital systems, cardiovascular with craniofacial systems, musculoskeletal with other BDs, and craniofacial with musculoskeletal systems. Relative to the Han, Zhuang had significantly lower risks in musculoskeletal  Table 4). The top six birth defect classes comprised over 85% of the total BDs (see Table 5).
The prevalence rates of top five BDs within musculoskeletal malformation, cardiovascular, craniofacial, urogenital, neurologic and other malformation were compared across five large ethnic groups (Table 6). With regards to musculoskeletal malformation, Zhuang, Miao, and Dong had lower prevalence of talipes equinovarus and Yao had a lower prevalence of pigmented nevus than Han. Three cardiovascular-specific risks (patent foramen ovale, patent ductus arteriosus and atrial septal defect) were less prevalent in Miao and Dong compared to Han. Congenital ear anomaly and genital anomaly were less prevalent in Zhuang, Miao, Dong, and/or Yao. Whereas hydrocephalus was more prevalent in Yao compared to Han. In addition, angioneoplasm was less prevalent in Zhuang and Miao and brain anomaly was less prevalent in Dong.

Associations of prevalence of BDs with perinatal characteristics
Results of the multivariate logistic regression analysis indicate that ethnicity was significantly associated with  (Table 7).

Discussion
Our study performed a comprehensive analysis of ethnic differences in the perinatal infant prevalence of BDs in Liuzhou, Guangxi, China over a 5 year time period. The average prevalence rate of total perinatal infants' BDs was 25.2 per 1000 PIs in the past 5 years, with an upward trend increasing by 1.49% on average each year. The ethnic distribution in our study sample was consistent with the ethnic distribution in Liuzhou, Guangxi as reported in the 2012 Chinese population census [21,23]. The average prevalence rate (25.2 per 1000 PIs) was higher than 20.2/1000, which was reported in a prior analysis of BDs in perinatal children in Guangxi from 2001 to 2010 [26]. The upward trend in the prevalence rate of BDs has been noted commonly in recent epidemiologic BD studies in China [3,18,[25][26][27][28]. The upward trend in BD epidemiology is potentially explained by several factors. Firstly, there is continued development of laws and regulations in Maternal and Child Health Care that have led to increased focus on BD surveillance and there is now full development of a government-led online BD surveillance system for the entire country. There is continued improvement in the overall health awareness of the country along with significant national financial support used in projects targeting tertiary prevention against BDs (pre-marital medical examination, pre-pregnancy care, genetic counselling, family planning, education on optimal reproductive age, and folic acid supplementation), along with the wider use of prenatal diagnosis techniques and the timely and effective diagnosis (e.g., Thalassemia carrier screening in pre-marital and pre-pregnancy check-up; primary health care ultrasound screening; nationwide newborn screening for congenital hypothyroidism, G6PD deficiency, phenylketonuria and galactosemia), and treatment and rehabilitation of the children who are born with defects (surgical operation for cleft lip with/ without palate and congenital heart disease) [3].
Our average prevalence rate was the highest among similar studies in various regions of China, such as Hunan  [18,25,29]. Our rate was lower than that reported from the United States (29.2 per 1000 PIs) in 2008, and the reported from the Korea (548.3 per 10,000 births, from 2009 to 2010), but higher than the reported from BDs in southern Vietnam (60.2 per 10,000 live births) [30][31][32]. Actually Hunan, Inner Mongolia and most of the other similar studies in China use a BD monitoring period between 28 weeks of gestation and 7 days after birth; [25,29] but in the United State the live birth hospitalizations are used in the 2008 Nationwide Inpatient Sample [30,31]. Li et al. found that the prevalence rate of BDs rose to 291.4 per 10,000 births from 244.2 per 10,000 births after including less than 28 week BDs in analysis of Guangdong BDs in 2007 [28]. Similarly, a study by Wang et al. showed that in a hospital-based survey from the prenatal 20th gestational week to postnatal 7 days, the prevalence of BDs was 232.7 per 10,000 births, which was significantly lower than that from the population-based survey (232.7 vs. 347.4 per 10,000 births, P < 0.001) [27]. In summary, the survey monitoring period and survey sample (hospital-based vs. population-based survey) did impact the prevalence rate and the raw number of BDs epidemiology.
The causes of BDs are complex, including a variety of risk factors such as: advanced maternal age, exposures   [29]. In several studies from the United States there were observed racial differences in the risk of BDs which were postulated to be related to genetic susceptibilities, cultural or social experiences that could modify exposures, or the combinations of genetic susceptibilities and environmental exposures [30,[34][35][36]. Does this mean ethnic groups in China have different genetic and partially-genetic causes of BDs or that the BDs could be due to their different life-style and environment exposures? Further epidemiologic study will be required to answer these questions.  CI confidence interval, RR relative risk, Mus Musculoskeletal, Cardio Cardiovascular, Cran Craniofacial, Uro Urogenital, gensyn genetic syndrome, resp respiratory a P < 0.05, b P < 0.01, c P < 0.001 Table 6 Highest five prevalences of specific birth defects from the No.

RR (CI)
Polydactylism or Symphysodactylia   Most previous Chinese studies showed that prevalence rates of overall BDs were different between rural and urban environments. For example, in the Inner Mongolia study, in the analysis of BDs in the Xinjiang multi-ethnic region, and in the government BDs Prevention Report in 2012, all concluded that a higher incidence of BDs in rural areas than urban areas [3,29,37].
However, in the study conducted in Guangxi from 2001 to 2010 in the epidemiology of BDs based on a BDs surveillance system from 2005 to 2014 in Hunan and the analysis of BDs in Hunan 2009, the prevalence of overall BDs in urban areas was significantly higher than that in rural areas [25,26,38]. These differences in results likely relate to socio-economic status, level of education, access to health care, exposure to environmental pollution and the life pressures which may be different between urban and rural areas. These factors may have direct or indirect impact on maternal exposures and could impact the prevalence rate of BDs [18,29,38,39]. Future studies from our center will attempt to correlate more information on residential location of the perinatal infant and any relation to ethnicity as a risk factor for the development of BDs.
The top six BD categories covered most BDs (86.8%) in our study. In a study of BDs in Guangxi from 2001 to 2010, the top six BDs were Schridde syndrome, polydactylism, congenital heart disease (CDH), total cleft lip, ear anomaly and hypospadias. This contrasted to the top six in our study which included musculoskeletal BDs as number one, and then cardiovascular, craniofacial, urogenital, other BDs, and neurologic BDs [26].
Thalassemia gene carriers in Guangxi comprise more than 20% of the population [7], which was the major cause of Schridde syndrome. This is a type of haemoglobin-hemolytic disease and was included in other BDs in our analysis. The most serious Schridde syndromes were almost all electively terminated before gestations of 28 weeks in recent years as a result of improvement in the premarital medical examination and at the level of prenatal diagnosis in Guangxi. Even as our study included gestation less 28 weeks, it still resulted in a lower prevalence of Schridde syndrome (15.27 vs. 29.57, per 10,000 PIs) compared to 2001-2010 in Guangxi [26].
The prevalence of CDH was increasing in our report, which is consistent with outcomes of other current studies from within China; and this would likely be due to the use of fetal ultrasound in antenatal care examinations and the improved surveying system online in China supported by the national finances in Maternal and Child Health Care System [3,18,28,40].

Conclusions
Our study found ethnic differences in the overall risk of BDs and types of BDs. To our knowledge, this is the first report of ethnic differences in the epidemiology of BDs in a hospital-based surveying system with internal quality controls in Guangxi, China. Our study provided a comprehensive description of ethnic differences in the risk of BDs in the most representative city (e.g., geography, composition of population) in Guangxi.