Gestational Diabetes Mellitus Among Asians – A Systematic Review From a Population Health Perspective

Objective Since Asians are particularly vulnerable to the risk of gestational diabetes mellitus (GDM), the lifecourse health implications of which are far beyond pregnancy, we aimed to summarize the literature to understand the research gaps on current GDM research among Asians. Methods We systematically searched the articles in PubMed, Web of Science, Embase, and Scopus by 30 June 2021 with keywords applied on three topics, namely “GDM prevalence in Asians”, “GDM and maternal health outcomes in Asians”, and “GDM and offspring health outcomes in Asians”. Results We observed that Asian women (natives and immigrants) are at the highest risk of developing GDM and subsequent progression to type 2 diabetes among all populations. Children born to GDM-complicated pregnancies had a higher risk of macrosomia and congenital anomalies (i.e. heart, kidney and urinary tract) at birth and greater adiposity later in life. Conclusion This review summarized various determinants underlying the conversion between GDM and long-term health outcomes in Asian women, and it might shed light on efforts to prevent GDM and improve the lifecourse health in Asians from a public health perspective. Systematic Review Registration Prospero, CRD42021286075.


INTRODUCTION
Diabetes is a significant cause of morbidity, mortality, and healthcare costs worldwide (1). The global age-adjusted comparative prevalence of diabetes among adults between 20-79 years of age was estimated at 8.3% (463 million) in 2019 (2), including 223 million women living with diabetes. And it is projected to reach 700 million people and 343 million women alone in 2045, respectively (2). Diabetes in pregnancy is similarly increasing in prevalence, with concerning consequences for both mother and offspring (3). Approximately 1 in 6 live births is affected by diabetes in pregnancy, 84% of which are diagnosed as gestational diabetes mellitus (GDM) (2,4).
GDM is defined as glucose intolerance with the first onset or recognition during pregnancy (2,4). Women with GDM have higher risks of cardiometabolic disorders during pregnancy and later in life (5). At the same time, offspring born to women with a history of GDM also encounter increased risks of developing obesity and other cardiometabolic disorders later in life (6,7). The documented prevalence of GDM varies substantially worldwide, ranging from 1% to >30% (3), while compelling evidence has shown Asians share a high prevalence (i.e., Middle East: 8.8-20.0%; South-East Asia: 9.6-18.3%; Western Pacific: 4.5-20.3%) (3) regardless of the racial/ethnic differences in body mass index (BMI).
A meta-analysis found a more than sevenfold increased risk of T2DM in women with GDM after index pregnancy, compared with women with normoglycaemic pregnancies (8). Data on risk factors -particularly modifiable risk factors that may inform effective intervention strategies are relatively more collected in the Western population (e.g., North America, Europe, and Oceania) than the Asian population (3,(8)(9)(10). Research reporting a full spectrum of long-term health outcomes among both mothers and offspring following pregnancies complicated by GDM also mainly stemmed from the Western population (11). Furthermore, GDM studies have not been comprehensively reviewed on Asian immigrants exclusively, given that an increasing number of Asian migrants live in Western countries for a long-term residency (12). Due to the different environmental exposures such as socioeconomic transitions, lifestyle adaptations, cultural assimilation hardship, and health disparities 9,10 , there might be exceptionally high attributable risks on GDM development for Asian immigrants compared with Native Asians.
This review sought to summarize the literature to understand research gaps and develop future research directions on Asian women with GDM from a population health perspective. Thus, our review serves the objectives to 1) comprehensively examine the epidemiology of GDM, its risk factors, and health consequences; and 2) identify areas for future research for public health interventions to prevent GDM and its health consequences.

Search Strategy and Selection Criteria
We conducted the systematic review according to PRISMA for systematic review protocols. References for this review were identified through searches of Pubmed, Web of Science, Embase, and Scopus for articles published until 30 June 2021. We included three topics in our review, namely "Topic 1-GDM prevalence in Asians", "Topic 2-GDM and maternal health outcomes in Asians", and "Topic 3-GDM and offspring health outcomes in Asians". Search terms included "prevalence", "incidence", "gestational diabetes mellitus", "gestational diabetes" and "diabetes in pregnancy" in combination with the terms "Asia", "Asians" and "Asian countries" in Topic 1. Search terms included "gestational diabetes mellitus", "gestational diabetes" and "diabetes in pregnancy" in combination with the terms "Type 2 diabetes", "prediabetes", "glucose intolerance", "abnormal glucose", "hypertension", "high blood pressure", "cardiovascular disease", "kidney disease", "cancer", "liver dysfunction", "non-alcoholic fatty liver disease" and "health outcomes" and also in combination with the terms "After delivery" and "postpartum" in Topic 2. Search terms included "gestational diabetes mellitus", "gestational diabetes", "diabetes in pregnancy" and in combination with terms "cardio-metabolic outcome", "cognitive outcome", "congenital disease", "adiposity", "hypertension", "health outcome", "neurocognitive outcome", "obesity", "diabetes", "cardiovascular disease", "kidney disease" and "cancer" and also in combination with "child" and "offspring" in Topic 3. Articles resulting from these searches and relevant references cited in those articles were reviewed, among which reporting non-Asian human subjects or without full-text available were excluded. Flow charts for literature searching on each topic are shown in Supplementary Figures 1-3. The Prospero registration number for this systematic review is registered as CRD42021286075.

Data Screening & Assessments
Double literature screening was conducted during the literature searching phase by two investigators (H L & L-J L). Furthermore, one investigator (A C) performed the quality assessments for all papers based on the Newcastle-Ottawa Scale Criteria (NOSC), and the other investigators (L-J L) verified the findingsindependently. The maximum score of 9 points in the Newcastle-Ottawa Scale is distributed in three aspects, namely selection of study groups (four points), comparability of groups (two points), and ascertainment of exposure and outcomes (three points) for case-control and cohort studies (13). We used the points to further categorize the publication quality with low risk of bias (between 7-9 points), high risk of bias (between 4-6 points), and very high risk of bias (between 0-3 points) (Supplementary Tables 1, 2). commonly used. Some countries adopted international guidelines as their national guidelines [e.g., China MOH guidelines (18), Malaysia MOH guidelines (19)], while some countries defined their own [e.g., Japan [Japan Diabetes Society] (20), India [Diabetes in Pregnancy Study group of India; DIPSI] (21), Turkmenistan (22), Oman (23)]. As the majority (123 out of 147) of included studies were published since 2010, we were not able to tease out whether the increment in GDM prevalence over the years in Asians is due to emerging evidence or new adoption of universal screening [i.e., IADPSG (14)].
We included studies using either one-step or two-step diagnostic guidelines, the latter of which performed a 1-h 50-g glucose challenge test (GCT) glucose challenging test (GCT) additionally during 24-28 weeks of gestation, with a whole blood glucose threshold of 7.2 mmol/l (130 mg/dl). In general, we observed a link between adopting any one-step diagnostic guidelines (e.g., the IADPSG guidelines, the WHO 1999 guidelines) and higher GDM prevalence among Asian studies.
For example, countries exclusively using (e.g., Singapore, UAE) or primarily using (e.g., China, Saudi Arabia, India) a one-step diagnostic approach reported an overall GDM prevalence above 10%. In contrast, countries exclusively using (e.g., Pakistan, South Korea) or primarily using (e.g., Thailand, Turkey, Japan) a two-step diagnostic approach reported an overall GDM prevalence below 10% ( Figure 3).

Prevalence of GDM in Asian Migrants
Twenty-eight studies reported GDM prevalence among Asian migrants in Europe, Oceania, and North America, with sample sizes ranging from 1,491 to 10,823,924 participants. Overall GDM prevalence among Asian migrants is comparable to the Native Asian population. However, the prevalence of GDM was generally higher in Asian immigrants (0.18%-24.2%) than non-Hispanic White (NHW) (0.02%-7.0%) living in the same country, regardless of GDM diagnostic guidelines used (Supplementary Table 5). Among Asian immigrants in UK and Norway, South, East, and West Asian immigrants, as a whole, had doubled the odds for GDM than NHW (24,25). Interestingly, length of immigration and birth countries seemed to relate to GDM prevalence. For instance, Danish-Chinese migrants with a longer stay (≥ 10 years) had a 62% higher odds of GDM onset than those with a shorter stay (≤ 5 years) (26). Also, foreign-born US-Indian migrants had a higher GDM prevalence than local-born US-Indian migrants (22.9% vs. 12.8%) (27).

Adverse Health Outcomes and Attributable Risk Factors Following an Index GDM-Complicated pregnancy
Overview Overall, seventy-two studies, predominantly longitudinal cohorts on GDM and maternal postpartum health outcomes, were identified in Asian countries ( Table 1 and Figure 4). Among them, prediabetes and T2D, cardiovascular disorders, cancer, and non-alcoholic fatty liver disease (NAFLD) were reported following index pregnancy complicated by GDM, with a mean or median follow-up from 4 weeks to 38 years after delivery. The majority of studies were reported from East Asia (42/72 studies, 58.3%), especially in the Chinese population. Two studies that reported postpartum T2D development in Asian immigrants were identified (Supplementary Table 6). Thirteen out of 74 included studies (18%) were assessed low in risk of bias, while the rest majority (80%) were either high or very high risk of bias (Supplementary Table 1).

Prediabetes and T2D
It is well-known that women with a history of GDM have a substantially increased risk of developing T2D than counterparts without such a history (8). A systematic review and metaanalysis on prospective studies with reasonable retention rates (mainly on European women) suggested that the conversion rate from GDM to T2D was seven folds increased among women GDM after index pregnancy, compared with those who had a normoglycaemic pregnancy (RR 7.43, 95% CI: 4.79-11.51) (8).
Sixty-three studies described the postpartum incidence rate of prediabetes and T2D among mothers diagnosed with GDM in Asia, with sample sizes ranging from 35 to 11 270 subjects, most of which defined prediabetes and T2D using the same guidelines [e.g., WHO 1999 (41) or ADA 2014 guidelines (42)] even though their GDM diagnostic criteria differed. We reported the percentage incidence (%) if prediabetes or T2D was recorded within one year from delivery (mostly between 6 and 12 weeks). Then we reported person-years incidence (per 1000 personyears) if prediabetes or T2D was recorded beyond one year from delivery (up to 15 years).
Within one year from delivery, the conversion rate varied significantly between studies from GDM to prediabetes (11.9%-49.1%) and from GDM to T2D (1.1%-66.7%), respectively. Beyond one year after delivery, the incidence rate from GDM to T2D was the highest in South Asia (47 -271 per 1000 personyears), followed by East Asia (9 -110 per 1000 person-years). We noted inconsistencies with study estimates within the same region. For instance, one study in Iran reported a much higher incidence T2D conversion rate than another study in Iran (172 vs. 9 per 1000 person-years) (35,43). Potential reasons for inconsistencies in the conversion rates from GDM to T2D could be the variation in studied population characteristics, duration of follow-up, retention rate, and data collection quality.
As for Asian immigrants, we identified only two reports comparing Asian immigrants with non-Asian counterparts, one from Spain with one-year follow-up (44) and the other from the US with an average 7.6-year follow-up (45). Both studies suggested that prediabetes and T2D conversion rates were higher in South Asian migrants than native NHW [prediabetes: 43.3% vs. 28.5% (44); T2D: 55 vs. 40 per 1000 person-years (45)].
Existing data on risk factors of T2D among women with a history of GDM were firstly reported in the NHW population,      (2) reduced high-density lipoprotein cholesterol <40 mg/ dL (1.03 mmol/L) in men or <50 mg/dL (1.29 mmol/L) in women or using specific treatment for this lipid abnormality; (3) raised blood pressure (systolic ≥130 mmHg or diastolic ≥85 mmHg or using antihypertensive drugs); and (4) raised fasting plasma glucose >100 mg/dL (5.6 mmol/L) or previously diagnosed type 2 diabetes. NCEP ATPIII criteria: MetS was defined if had at least three of the following: (1) waist circumference ≥90 cm in men, or ≥80 cm in women; (2) systolic blood pressure ≥130 mmHg, and/or diastolic blood pressure ≥85 mmHg, or using antihypertensive drug treatment; (3) fasting glucose ≥100 mg/dL, or using drug treatment for elevated glucose; (4) triglyceride ≥150 mg/dL or using drug treatment for elevated triglycerides; (5) high-density lipoprotein cholesterol <50 mg/dL in women, or <40 mg/ dL in men, or using drug treatment for reduced high-density lipoprotein cholesterol. AHA/NHLBI criteria: MetS was defined if 3 out of the following 5 criteria are met, (1) waist circumference>80 cm, (2) blood pressure >130/85 mmHg or on antihypertensive medication, (3) fasting plasma glucose >100 mg/dL or on antidiabetic medication, (4) fasting triglyceride >150 mg/dL, (5) high-density lipoprotein <50 mg/dL or on antihyperlipidemic medications. 2007 ESH-ESC Guidelines: hypertension was defined as systolic blood pressure ≥ 140mmHg or diastolic blood pressure ≥ 90 mmHg or taking antihypertensive medicines *Incidence rate in per 100 000 person year is only calculated when the mean year of follow-up is above 1 year.
such as greater pre-pregnancy BMI (8,9), excessive weight gain (3), unhealthy dietary patterns (3), physical inactivity (3), and a short period of lactation (3,10). In the Asian population, there are also quite a few at-risk pre-natal maternal characteristics recently added to this pond of evidence, such as family history of diabetes (43), a higher degree of consanguineous marraiges (43), higher pre-pregnancy BMI (29,31,32,46), higher total cholesterol quartile at GDM diagnosis during the index pregnancy (47), younger age at delivery (<30 years) (46), and a short period of lactation (<6 months) (33). Post-natal risk such as missing medical assistance in the continuum of GDM care after delivery could be another risk for T2D progression among Asian mothers with a history of GDM (48).

Cardiovascular Disorders Hypertension
A history of GDM was related to increased risk of hypertension (HTN) after the index pregnancy in some but not all studies. For instance, the US Nurses' Health Study found an increased risk of postpartum HTN among women with a history of GDM (49). In contrast, a Dutch cohort suggested the risk of developing HTN was mainly significant among women with a history of hypertensive disorders during pregnancy (HDP) rather than GDM (50). Among the three studies identified in our review on GDM and subsequent hypertension risk (28,30,38), the Chinese Tianjin GDM prevention program reported a much higher incidence rate of HTN among women diagnosed with HDP and GDM than women with GDM alone (118 vs. 26 per 1000 person-years) (38), which partially agreed with the Dutch cohort.
The mechanisms underlying postpartum HTN in women with GDM remain un-elucidated. Insulin resistance may be a component of the underlying pathophysiology linking GDM with postpartum HTN, with or without HDP (51). As we know, obesity and excessive weight gain during pregnancy are associated with insulin resistance (38), inflammation and oxidation (52), all of which may lead to permanent vascular damage (51) and even irreversible peripheral vascular resistance. Due to the largely inadequate evidence, future research to investigate the role of antenatal and postpartum lifestyle (e.g., dietary patterns, physical activities) in the progression of HTN is warranted in Asians.

Cardiovascular Risks and Cardiovascular Diseases
Emerging evidence has led to the increasing recognition of the association between GDM and cardiovascular (CV) risks and CV events later in life (53). Previous studies in the Western population have identified a higher level of inflammatory (e.g., C-reactive protein) (54), vascular endothelial dysfunction (e.g., intimal medial thickness) (55), and a 2-7 times higher risk of coronary artery calcification or CVD after 12-15 years' follow-up (56)(57)(58), among women with a history of GDM. In Asia, five studies reported metabolic syndrome in Asian women with a history of GDM, with an incidence rate ranging from 40 to 90 per 1000 person-years. One Chinese study reported postpartum dyslipidemia (38.5%) among women with a history of GDM (47), while the other Israelite study reported a 30-70% higher risk of developing CV events and CV hospitalization among women with a history of GDM, even after adjusting for pre-eclampsia and maternal obesity at index pregnance (39).
Thus far, only determinants for postpartum CVD risks and CV events were reported as family history of T2D (59) and postpartum development of T2D (58) in the western population. Even though postpartum CVD determinants among women with GDM have yet to be fully investigated, long-standing exposure to cardio-metabolic risks has been speculated in the GDM-CVD link.
Cancer GDM was associated with 30-40% increased risks of breast cancer, thyroid cancer, stomach cancer, and liver cancer for all races and ethnicities in a recent meta-analysis (60). As in the Asian population alone, we identified six retrospective cohort studies (Taiwan, South Korea and Israel) using either national insurance or a medical database to investigate the association between GDM and various cancers. All of them reported higher incidences of breast cancer, thyroid cancer, pancreatic cancer, ovarian cancer, lung cancer, and kidney cancer among the Asian female population with a history of GDM after a median of 5-38 years of follow-up than those parous women without such a history. For example, the incidence rate of cancer among Israelite women with a history of GDM was reported in breast (2 per 1000 person year) (37) and ovary (1 per 100 person year) (36), respectively.
It has been well documented that T2D is associated with higher risks of all-cancer incidence (61), especially malignancies in the breast, pancreas, and liver in women (62,63). Some evidence has alluded to the mitogenic effect while binding to the insulin-like growth factor-I receptor secondary to insulin resistance (64). Furthermore, hyperglycemia itself might promote carcinogenesis via increasing oxidative stress (65,66). However, data regarding cancer risks associated with GDM are merely gathered in the Western population.

Liver Dysfunction
Liver dysfunction is a common cause of chronic liver disease that affects approximately one in four adults worldwide, which is characterized by liver steatosis (fat deposition), inflammation, and hepatocyte damage (67). Researchers have suggested a link between metabolic risks (i.e., obesity, hyperglycemia, hyperlipidemia, and insulin resistance) and hepatic fatty deposition and non-alcoholic fatty liver disease (NAFLD) in the past decades (68,69). Notably, women with a history of GDM were found to have raised liver triglyceride (TG) levels, highlighting a potential link between GDM and liver dysfunction (70,71). Despite the higher prevalence of postpartum liver fat (72), abnormal liver score (73) and even NAFLD (71,74), such results were mostly gathered from the Western population. There is one study from South Asia (India) reported a 2.11fold higher odds of NAFLD among women with GDM, compared with women without GDM. The researchers suggested that postpartum medical conditions such as overweight/obesity, metabolic syndrome, and prediabetes were risk factors for developing NAFLD, during a median of 16 months' follow-up after delivery (40).

Adverse Health Outcomes of Offspring Born From Pregnancies Complicated by GDM
Overview A body of evidence has implied that specific developmental programming in offspring is influenced by maternal hyperglycemia; in particular, epigenetic modification may be the key underlying mechanism (75,76). Our review identified forty-two studies conducted on Native Asians ( Table 2) and eight studies conducted on Asian immigrants (Supplementary Table 7) with up to 18 years' follow-up, all of which were within the research scope of adverse health outcomes among offspring born to mothers with GDM. Offspring health outcomes, including fetal growth and neonatal anthropometric measures, were reported in Native Asians and Asian migrants, whereas offspring health outcomes, including congenital anomalies, neuro-cognitive function, and cardio-metabolic phenotypes, were only reported in Native Asians ( Figure 4). None of these studies investigated risk factors underlying maternal GDM and the development of offspring health outcomes. Among 50 included studies in this topic, fourteen (28%) were assessed low in risk of bias, while the rest 72% were assessed either high or very high in risk of bias.

GDM and Fetal Growth
In-utero over nourishment can lead to fetal overgrowth, and such influence may predispose the offspring to obesity and T2D later in life if there is an obesogenic environment (84). A cohort in India reported an association between GDM and antenatal fetal growth at mid-late trimester (85). In this prospective cohort, fetuses of women with GDM had a thicker anterior abdominal wall while smaller femur length and biparietal diameter than fetuses of women without GDM. The researcher referred to this as "the thin-fat-phenotype" which represented a predisposition to T2D at birth (85).
Among Asian immigrants, one Norwegian study found that fetuses exposed to maternal GDM tended to be smaller in fetal weight at 24 weeks of gestation but thereafter grew faster until delivery, compared with fetuses not exposed to maternal with GDM (86). This trend was more prominent in South Asian women (86).

GDM and Neonatal Outcomes Anthropometric Outcome At Birth
It is well-accepted that GDM is related to increased risk for macrosomia and large for gestational age (LGA) (6). We identified 14 papers that focused on this topic, with sample sizes ranging from 72 to 11 999 neonates. Among them, the majority reported consistent findings on either higher prevalence rates (11% to 40%) or higher risk ratios (2.0-2.7 times) of macrosomia or LGA among neonates born to GDM mothers, compared with their non-GDM counterparts, despite a couple reported otherwise. Interestingly, one study specifically looked at different combinations of glycemic abnormalities (fasting, 1hour, and 2-hour glycemic levels) with macrosomia (77). The researchers found that women with three abnormal OGTT glycemic values had a much higher macrosomia rate in their offspring than those with two or one abnormal glycemic value (77). Such results-to some extent-suggested there might be remarkable neonatal outcomes specific to different GDM phenotypes (77).
Four studies reported neonatal birth size in Asian migrants equivocally. The US studies showed no differences in macrosomia rate between neonates born to NHW and Asian women with GDM (87,88). In contrast, compared with the NHW counterparts, the Dutch study showed a lower macrosomia rate in offspring born to West Asian migrants (Turkish) (

Neonatal Health Ouctomes
Eight papers reporting other neonatal conditions were identified in our review, ranging from 72 to 10 543 in sample size. Neonatal disorders were listed as hypoglycemia, low Apgar score, hyperbilirubinemia/jaundice, polycythemia and respiratory distress syndrome. All studies consistently reported that neonates born to women with GDM were more susceptible to hypoglycemia, hyperbilirubinemia, respiratory distress syndrome and low Apgar score (<7 at 5 minutes), compared with those born to women without GDM.

Congenital Diseases
A total number of six studies reported findings on this topic, only half of which had specified the type of malformation as either congenital heart disease or congenital anomalies of the kidney and urinary tract (CAKUT). In general, evidence showed that neonates born to mothers with GDM tended to have a 2-3 times higher risk of developing congenital heart disease and CAKUT, especially more evident in male neonates (79). Despite the unclear pathophysiological mechanism, it has been speculated that serial maternal antenatal characteristics could affect embryonic development during the first trimester, such as preexisting diabetes prior to pregnancy, overweight and obesity, and excessive weight gain during pregnancy (79,91,92).

Neuro-Cognitive Structure and Function
There is one case-control study investigated brain function in pre-term infants born to mother with GDM. In the first 33 days after delivery, the researchers used MRI image and discovered that infants born to mother with GDM tended to have multiple reduced fractional anisotropy in the brain, reflecting a microstructural white matter abnormalities compared with the infants born to mother without GDM (80).

GDM and Childhood Outcomes
Twenty studies on this topic were identified, with nearly half reported in China (n=8), then followed by India (n=4), Israel (n=3), Hong Kong (n=3), Pakistan (n=1), and Sri Lanka (n=1). Childhood outcomes spanned several traits and conditions, including adiposity and cadiometabolic outcomes, cognitive function, endocrinological and ophthalmological morbidity.

Anthropometry, Blood Pressure and Cardiometaboilc Outcomes
The majority of studies (17/20, 85.0%) reported consistent findings on long-term outcomes like childhood adiposity and cardiometabolic risks. Overall, offspring born to women with GDM had higher BMI z-score, higher systolic blood pressure and diastolic blood pressure, higher childhood overweight and obesity rates, higher lipid profile levels, and higher insulin and insulin resistance levels, than those born to women without GDM. These studies involved small (n=164) to large (n=27 157) sample sizes of offspring with an average follow-up of 1-18 years among different ethnicities (Chinese, Indians, Sri Lankans and Israelite Jews).
In terms of cardiac function, we included one Pakistani study (93) and one Indian study (81) with small sample sizes of 136 and 236. Compared with their counterparts, offspring born to women with GDM had higher Carotid Intima-Media Thickness (cIMT), cardiac output and stroke volume, decreased mitral E/A ratio, and total peripheral resistance in early childhood and early adolescence, respectively. Among Asian immigrants, two studies in the UK (94, 95) and one study in the US (96) with sample sizes ranging from 382 to 6 060 reported a consistent association between GDM and childhood obesity across all races and ethnic groups. The magnitude in such association between NHW women and Asian female immigrants was similar.

Neuro-Cognitive Outcomes
Hyperglycemia during pregnancy may affect fetal neurodevelopment and leave a significant impact on offspring cognition (97). Only one Indian study reported neurocognitive outcomes in the offspring at a mean 9.7 years of age (82). Children born to women with GDM had higher learning, longterm retrieval and storage, and better verbal ability than children born to women without GDM. The authors propose that the finding may be confounded by the strong correlation between GDM and higher social-economic status among this cohort (82).

Endocrinological and Ophthalmological Outcomes
Other childhood outcomes related to GDM include endocrine and ophthalmic morbidities. In two large-scale Israelite cohort studies where young adults (≤ 18 years) with a history of smallthan-gestational age (SGA) conditions were recruited. One study showed no difference in the incidence of endocrine morbidity between young adults born to women with and without GDM (83). In contrast, the other study observed a higher prevalence of offspring ophthalmic inflammation (0.74% vs. 0.60%) and a 60% higher risk in ophthalmic-related hospitalization among young adults born to women with GDM and treated with medication (metformin, insulin) (78).

DISCUSSION AND FUTURE DIRECTION
Our review reinforces that, in general, Asians are at the highest risk of developing GDM and for subsequent progression to T2D among all populations. Yet, data among the Asian population on long-term health implications of GDM on women and offspring remain limited and are less in-depth than the Western population. In addition, studies in identifying attributable risk factors that may inform preventive strategies of long-term adverse health outcomes among women and their offspring are less comprehensive in Asians than in the Western population. Methodologically, inferences from existing published data are hindered by considerable heterogeneity in study designs, a high risk of bias (Supplementary Tables 1, 2), and standardized protocols for defining studies of Asians.
In order to address such critical knowledge gaps, future endeavors in the following aspects may be warranted to dissect          GDM, gestational diabetes mellitus; DM, diabetes mellitus; HC, head circumference; AC, abdomen circumference; FL, femur length; BPD, biparietal diameter; BMI, body mass index; LGA, large for gestational age; OR, odds ratios; OGTT, oral glucose tolerance test; CAKUT, congenital anomalies of the kidney and urinary tract; SD, standard deviation; HR, hazard ratio; BP, blood pressure; cIMT, carotid intima media thickness.
the vicious circle of "diabetes begetting diabetes" and improve the health and well-being of this and future generations. 1. Conducting large scale well-designed cohort studies and/or consortium networks among Asians to investigate risk factors and etiology of GDM. A better understanding of GDM pathogenesis specific to Asian women shall further enhance our knowledge on the unique GDM characteristics among Asian women and develop more targeted and effective intervention approaches to prevent GDM and interrupt the transgenerational diabetic vicious cycle. However, such GDM heterogeneity-specific maternal health outcomes in Asians are still limited in scope, let alone other elements of the potential impact such as genetic factors and fetal sex. Future endeavors to establish parallel prospective pregnancy cohorts-with longitudinal data collection and comprehensive characterization of metabolic profiles through pregnancy in different Asian regions-are warranted to understand biological differences across Asian ethnicities, identify determinants and even develop prediction models for GDM onset and its phenotypespecific transgenerational health outcomes.
2. Conducting prospective cohort studies and/or intervention studies to follow up both GDM women and their offspring following the index pregnancy to identify factors that may mitigate the adverse impact of GDM on both women and their children. With the increasing awareness of the GDM burden and subsequent adverse health outcomes in Asian women and their offspring, a few large-scale ongoing pre-conception and pregnancy trials have focused on lifestyle intervention in Asia, such as Project SARAS in Mumbai (98) and the VINAVAC study in Vietnam (99). However, inferences from these two trials are inconsistent, which might be hindered by participants' low compliance, including low uptake rate of OGTT, poor quality of data collection (e.g., physical examination, questionnaires administration) during research visits, and not quantitative constituents in the snack or freshly-prepared food given to the intervention group (98,99). In terms of postpartum trials, substantial evidence in either lifestyle modifications (100) or pharmacological therapies (101-103) gathered from developed countries has shown promising results. However, intervention studies with customized approaches (e.g., diet recommendation, lifestyle modification) according to the Asian population are much fewer in scope than the Western population. Recently, there have been some improvements, including a few postpartum T2D prevention trials conducted in countries like China (100, 104), Singapore (105), Malaysia (106), and India (107), focusing on lifestyle modification, with a sample range between 77 and 1 414 and a length of follow-up up to 10 years. However, most of them are still ongoing, and only two trials reported more significant weight loss, reduction in waist circumference, and improved glucose tolerance during the 6-12 months' postpartum period (104,106).
3. Conducting studies of Health Disparities in GDM Care in Asian Populations across countries and continents. Even though developing countries in Asia (e.g., India) have shown increased life expectancy over the past several decades, health inequity is still a severe national issue as progress is uneven within each country (108). Furthermore, not all but a substantial proportion of Asian migrants in Western countries face socio-economical disadvantages such as access to health care and education (109). Among them, women seem to be more affected than men due to their vulnerability (109). Therefore, the fight against GDM and its harm to Asian mothers and children should account for existing health inequity and develop strategies to address health disparities.
4. Health Care System Improvement in Asia. Emerging evidence has pointed out that a portion of GDM cases was indeed overt diabetes that has not been identified before pregnancy, which ultimately drives the risk of maternal and offspring health outcomes even higher (110). For example, collecting information on pre-existing maternal diabetes or overt diabetes identification during early pregnancy in the Asian health care system is critical to screen for and even prevent offspring congenital abnormality or other adverse fetal and neonatal health outcomes. Ideally, GDM rates in the population could be reduced by individual and societal measures designed to promote healthy lifestyle changes, including optimal dietary intake and increased physical activity in the general population, focusing on the health and fitness of women of reproductive age.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding authors.

AUTHOR CONTRIBUTIONS
L-JL contributed to the review's framework conceptualization, study, design, literature research, data collection, analysis and interpretation, and manuscript write-up; LH contributed to literature search, data collection and summary; DT contributed to data interpretation and manuscript editing; CZ contributed to the review's framework conceptualization, study design, data interpretation and manuscript editing. All authors contributed to the article and approved the submitted version.