ObjectiveTo study whether the current Institute of Medicine (IOM) pregnancy weight gain recommendationsvarybypre-pregnancybodymassindex(BMI)wassuitabletoChinese people.MethodsA study was conducted on 4736 term singleton live birth gravidas,who were diagnosed normal glucose metabolism and delivered in Peking University First Hospital in 2005 and 2009,by reviewing the medical records.Based on the pre-pregnant BMI,the selected cases were divided into 3 groups:low body mass group ( BMI ＜ 18.5 kg/m2,n =465 ),normal body mass group ( BMI 18.5 - 24.9 kg/m2,n =3549),over body mass group ( BMI ≥ 25 kg/m2,n =722).All the cases were divided into 3 subgroups based on pregnancy weight gain as below,within,and above the IOM recommendations in each pre-pregnant BMI group.Totally 4736 newborns were divided by birth weight into 3 groups:normal birth weight group ( weight 2500 - 4000 g,n =4339 ),macrosomia group ( weight ≥ 4000 g,n =359 ) and low birth weight group (weight ＜ 2500 g,n =38).The difference of age,gestational age,pre-pregnant weight,pre-pregnant BMI and history of delivery of cases between 2005 and 2009 were analyzed.The difference of pregnancy outcome of women whose gestational weight gain was below,within,and above the IOM recommendations was analyzed.Results (1) Compared to mothers with pregnancy weight gain within IOM recommendations in low body mass group,risk of low birth weight in offspring was elevated tendency with pregnancy weight gain below IOM recommendations ( OR =3.71,95％ CI:0.97 - 14.12,P =0.055 ).(2) In normal body mass group, compared to women with pregnancy weight gain within IOM recommendations, risk of macrosomia in offspring was elevated with pregnancy weight gain above IOM recommendations ( OR =2.14,95％ CI:1.62 - 2.83,P ＜ 0.01 ).( 3 ) In over body mass group,compared to women with pregnancy weight gain within IOM recommendations,risk of macrosomia in offspring was elevated ( OR =3.25,95％ CI:1.65 -6.39,P =0.001 ) and risk of hypertensive disorders complicating pregnancy was high ( OR =1.79,95％ CI:1.04 -3.09,P =0.037 ) in women with pregnancy weight gain above IOM recommendations.ConclusionThe current IOM pregnancy weight gain recommendations vary by pre-pregnancy BMI may be suitable to Chinese people.

Objective To investigate the relationship between gestational hyperglycemia and adverse pregnancy outcomes and find out the optimum diagnostic criteria of gestational diabetes mellitus in China. Methods A retrospective population-based study of 14 593 pregnant women, who delivered between Jan. 2005 and Dec. 2009 and accepted the gestational diabetes mellitus ( GDM ) screening and diagnosis was performed. The prevalence of gestational hyperglycemia according to different criteria was calculated, and the incidence of adverse pregnant outcomes relation to gestational hyperglycemia according to different criteria was analyzed. Results ( 1 ) According to National Diabetes Data Group (NDDG) criteria and International Association of Diabetes and Pregnancy Study Groups (IADPSG) criteria, the prevalence of gestational hyperglycemia that intervention required was 8.9% (1293/14 593 ) and 14.7% (2138/14 593 )respectively; the prevalence of gestational hyperglycemia differed significantly between NDDG and IADPSG criteria ( P ＜ 0. 05 ). ( 2 ) The prevalence of macrosomia, large for gestational ages ( LGA), cesarean section,preterm birth and neonatal hypoglycemia etc would increase in gestational glucose metabolic disorders according to any criteria. The prevalence of the complications in gestational hyperglycemia according to NDDG criteria, IADPSG criteria and the patients with normal glucose metabolism is as follows, macrosomia:8.4% ( 108/1293), 11.3% (241/2138) and 6. 7% ( 835/12 403 ); LGA: 9. 7% ( 125/1293 ), 11.7% (250/2138) and 5.5% (687/12 403); cesarean section: 59. 0% (763/1293), 60. 4% ( 1291/2138 ) and 51.6%(6397/12403); preterm birth: 11.4% (147/1293), 9.5% (203/2138) and 6.3% (777/12 403); neonatal hypoglycemia: 2. 6% ( 33/1293 ), 2. 2% (46/2138) and 0. 7% ( 89/12 403 ). ( 3 )About 71.3% (922/1293) of the gestational hyperglycemia according to NDDG criteria could be well control only by diet control. Conclusion The prevalence of perinatal complications would increase in gestational hyperglycemia that achieved IADPSG criteria without intervention, so IADPSG criteria is reasonable in China.

Objective To investigate the relationship between early pregnancy fasting plasma glucose (FPG) and gestational glucose metabolism disorders. Methods Six hundred and fifty-six pregnant women who were singleton, non-diabetes before pregnancy and had FPG examined during 5-13 weeks of pregnancy were admitted into this study from January 1, 2009 to May 31, 2009. All these subjects had routine prenatal examination and finally delivered in the Department of Obstetrics of Peking University First Hospital. The FPG, 50 g glucose challenge test (GCT) after 24 weeks of pregnancy, 75 g oral glucose tolerance test (OGTT), gestational diabetes mellitus (GDM),gestational impaired glucose tolerance (GIGT) were analyzed with receiver operating characteristic (ROC) curve. Results (1) Relationship between FPG and GCT were analyzed with ROC curve.The maximum area under curve was 0. 539 (95% CI: 0. 493-0. 586) and there was no correlation between the FPG and GCT results(P=0. 057). (2) Relationship between early pregnancy FPG and abnormal FPG examined after 24 gestational weeks were also analyzed . The maximum area under curve was 0. 796(95% CI: 0. 672-0. 920). If 5. 05 mmol/L was taken as the cutoff value, the sensitivity and specificity was 54. 5% and 83. 2%, respectively. There was significant relationship between the two values (r=0. 432, P=0. 000). (3) There were no relationship between early pregnancy FPG and the blood glucose value of 1, 2 and 3 h in 75 g OGTT (r=0. 093, 0. 036 and 0. 107, P=0.122, 0. 549 and 0. 074 respectively). OGTT 0 h value was positively related to OGTT 1, 2 and3 h glucose level (r=0.493, 0.421 and 0.368, P=0.000, respectively). (4) All early pregnant FPG values in this study were under 6.1 mmol/L. Twenty-two GDM and 27 GIGT patients were diagnosed in this study. Early pregnancy FPG did not relate to the GDM and GIGT diagnosis.Conclusions Early pregnancy FPG could not replace 50 g GCT as an early screening for glucose metabolic abnormality in pregnancy, but FPG during early pregnancy is necessary.

Objective To investigate the appropriate range of gestational weight gain in pregnant women with abnormal glucose metabolism.Methods A retrospective study was conducted on 661 term singleton pregnant women with gestational abnormal glucose metabolism,who delivered in the Department of Obstetrics and Gynecology of Peking University First Hospital from Jan.2005 to Dec.2007,by reviewing the medical records.All sujects were divided into 4 groups according to their body mass index (BMI) before pregnancy:group Ⅰ (n=40):BMI<18.5;group Ⅱ (n=400):BMI18.5-23.9;group Ⅲ (n=162):BMI 24.0-27.9;group Ⅳ (n=59):BMI≥28.0.The weight gain among different groups and that between women who delivered normal birth weight infant and maerosomia were analyzed.The weight gain of pregnant women who delivered babies weighing 3000～3500 g in each group was determined as the appropriate weight gain for that group.Results The same results were achieved that the weight gain in pregnant women who delivered macrosomia was significantly higher than those who delivered normal birth weight newborns in each group,ie,the weight gains for women who had macrosomia and normal birth weight infants were (17.0±5.2) kg and (14.1±4.7) kg in group Ⅱ,(16.8±7.3) kg and (11.9±5.1) kg in group Ⅲ and (18.3±6.7) kg and (11.2±5.4) kg in group Ⅳ,respectively (P<0.05).The appropriate ranges of weight gain for each group were (15.6±3.3) kg,(14.0-18.0) kg for group Ⅰ,(13.9±4.6) kg,(11.0-16.5) kg for group ]],(11.5±5.2)kg,(9.0-15.0) kgforgroup Ⅲ,(10.1±2.9) kg,(7.0-12.7) kg forgroup Ⅳ.Conclusions Appropriate weight gain based on prepregnant BMI,together with glucose monitoring in women with gestational abnormal glucose metabolism,is helpful for fetal weight control.

Objective To investigate the expression of glucose transporters(GLUTs)in placentas of pregnant women with abnormal glyeometabolism,and to explore its effect on glucose transport between mother and fetus and its relation with the birth weight.Methods Placentas of 41 pregnant women with abnormal glycometabolism(7 cases of DM,10 GDM A1,10 GIGT and 14 GDMA2)and 15 normal pregnant women as control were selected.The expression of GLUT1 and GLUT3was detected by immunohistochemistry.The birth weight was measured at delivery.Results GLUT1 was expressed in the syncytiotrophoblasts and cytotrophoblasts,whereas GLUT3 in some endothelial cells.The expressions of GLUT1 and GLUT3 were significantly different among the five groups(P＜0.01).Positive correlation was shown between the GLUT1 expression and the birth weight(rs=0.532,P＜0.01),but not in GLUT3 expression.Conclusions The expression of GLUT1 and GLUT3 in placentas of pregnancy with abnormal glycometabolism is enhanced,and GLUT1 may play a predominant role in the fetal glucose uptake.

Objective To study three single nucleotide polymorphisms (SNP), SNP-43, -19 and - 63 of calpain-10 (CAPN10) gene in pregnant women with glucose metabolism disorders and their relationship with the risk of these disorders. Methods Totally, 270 pregnant women including 156 with an abnormal oral glucose tolerance test (study group) and 114 healthy controls were selected among those delivered at the Department of Obstetrics and Gynecology, Peking University First Hospital from Jan. 2005 to Dec. 2006. The SNP of CAPN10 gene at posifons 43, 19, and 63 were analyzed by primer-introduced restriction analysis-polymerase chain reaction (PIRA-PCR). Results (1) The frequencies CAPN10 SNP-19 2R/2R genotype (26.9% ,42/156) and 2R allele (48.9%, 152/312 ) were higher than those in controls (12.3% ,14/114 and 36.8% ,84/228, respectively; P=0.012, 0.006). (2) Stratified analysis according to family history of diabetes mellitus showed the proportion of the CAPN10 SNP-19 2R/2R+2R/3R genotypes (83% ,47/57) in the study group were significantly higher than that of control group (52%,11/21 ; P=0.009), and the proportion of SNP-63 T/T + T/C genotypes(47% ,27/57) in study group were significantly higher than that of control group (14%, 3/21 ; P=0.026) among those with positive family history. (3) The combined effect of CAPN10 SNP-43, -19 and -63 on glucose metabolism disorders was analyzed by comparing with the other haplotypes and showed that the haplotype 121 distribution in study group was significantly higher than those in controls(P=0.036), and 221 haplotype was significantly lower than controls (P=0.042). Conclusions (1) CAPN10 SNP-19 is associated with glucose metabolism disorders in pregnant women. And 2R allele might be the risk factor. CAPN10 SNP-19 2R/2R +2R/3R and SNP-63 T/T + T/C genotypes might increase the risk of glucose metabolism disorders in women with positive family history. Haplotype 121 might increase the risk of glucose metabolism disorders in pregnant women and 221 be a protective factor.

Objective To investigate the effects of gestational diabetes mellitus (GDM) on the growth rate of fetuses.Methods This was a retrospective study.Women who had deliveries in Peking University First Hospital between January 2012 and June 2013 were enrolled.Matched by maternal age,they were divided into four groups with 100 cases in each group:macrosomic fetuses of mothers with GDM (GDM-macrosomia group),normal birth weight infants of mothers with GDM (GDM-non-macrosomia group),macrosomic fetuses of mothers with normal pregnancy (normal-macrosomia group),and normal birth weight infants of mothers with normal pregnancy (normal-non-macrosomia group).The fetal abdominal circumference was measured at 20+1-24,28+1-32,32+1-37 weeks and 37+1 weeks to delivery under prenatal ultrasound.The growth rate of fetal abdominal circumference was calculated (abdominal circumference/gestational weeks).The analysis of variance,the least significant difference-t test and Student's t test were used for statistical analysis.Results At early pregnancy [(9.2±2.6) weeks of gestation] and 20+1-24,28+1-32,32+1-37 weeks and 37+1 weeks to delivery,the weight and body mass index (BMI) of the mothers in GDM-macrosomia group were higher than those in the other three groups (all P＜0.05).The body weight of the mothers increased by (15.5±5.4),(13.5±3.6),(16.4±4.1) and (16.2±4.3) kg,respectively,compared with early pregnancy.At 20+1-24,28+1-32,32+1-37 weeks and 37+1 weeks to delivery,the fetal abdominal circumference of GDM-macrosomia group was (182.0± 13.9),(270.7± 17.7),(335.2±21.3) and (362.3± 18.7) mm,respectively,being higher than that in GDM-non-macrosomia group [(176.8± 13.0),(256.6± 13.5),(313.2± 17.5) and (335.8± 15.5) mm] and normal-nonmacrosomia group [(176.9± 11.8),(260.0± 14.2),(310.6± 21.4) and (334.5 ± 16.1) mm] (all P＜0.05).The fetal abdominal circumference of GDM-macrosomia group was even higher than normal-macrosomia group at 32+1-37 weeks [(335.2±21.3) vs (326.1 ± 19.1) mm,t=4.01,P＜0.05].The changes of the growth rate of fetal abdominal circumference were consistent with the fetal abdominal circumference.Conclusions GDM accelerates fetal growth.

Objective To compare the clinical use of continuous glucose monitoring system (CGMS) and self-monitoring blood glucose (SMBG) when monitoring blood glucose level of patients with gestational diabetes mellitus (GDM) or type 2 diabetes mellitus (DM) complicated with pregnancy.Methods A total of 99 patients with GDM (n=70) and type 2 DM complicated with pregnancy (n=29) that whether hospitalized or in clinical of Peking University First Hospital were recruited from Aug 2012 to Apr 2013.The CGMS was used to monitor their blood glucose level during the 72-hour time period,while the SMBG was also taken seven times daily.The correlation between these blood glucose levels and their glycosylated hemoglobin (HbA1c) levels were analyzed by comparing the average value,the maximum and the minimum value of blood glucose,and the appeared time of these extremum values in these two monitoring methods,and the amount of insulin usage was recorded as well.Results (1) The maximum,minimum and the average blood glucose value in the GDM group were (8.7± 1.2),(4.5 ±0.6) and (6.3 ± 0.6) mmol/L of SMBG vs.(10.1±1.7),(3.1±0.7),(6.0±0.6) mmol/L of CGMS.These values in DM group were (10.1±2.2),(4.5±1.0),(6.9±1.1) mmol/L of SMBG vs.(12.2±2.6),(2.8±0.8),(6.6±1.1) mmol/L of CGMS.By using the two methods,the maximum and the average value of the two groups showed significant differences (P＜0.01) while the minimum value showed no significant differences (P＞0.05).(2) In the GDM group,the average blood glucose values of CGMS and SMBG were significantly correlated (r=0.864,P＜0.01).The maximum values presented the same result (r=0.734,P＜0.01).Correlation was not found in the minimum values of CGMS and SMBG (r=0.138,P＞0.05).In the DM group,the average valves of two methods were significantly correlated (r=0.962,P＜0.01),the maximum values showed the same result (r=0.831,P＜0.01).It can also be observed in the minimum values (r=0.460,P＜0.05).(3) There was significant correlation between the average value of CGMS and HbA1c level (r=0.400,P＜0.01),and the average value of SMBG and HbA1c level were correlated (r=0.031,P＜0.05) in the GDM group; the average values of CGMS (r=0.695,P＜0.01) and SMBG (r=0.673,P＜0.01) were both significantly correlated with the HbA1c level in the DM group.(4) In the GDM group,37％ (26/70) of the minimum values of SMBG appeared 30 minutes before breakfast,while 34％(24/70) of them appeared 30 minutes before lunch; 86％(60/70) of the maximum values of SMBG were evenly distributed 2 hours after each of the three meals.In the DM group,41％(12/29) of the minimum values of SMBG presented 30 minutes before lunch,while 21％(6/29) and 14％(4/29) of them were showed 30 minutes before breakfast and dinner respectively; about 30％ of the maximum values of SMBG appeared 2 hours after each of the three meals.(5) In the GDM group,23％(16/70) of the minimum values of CGMS occurred between 0:00-2:59 am.,and most of the other minimum values of CGMS were evenly distributed in the rest of the day,except for 3％(2/70) of them were found during 18:00-20:59 pm.43％(30/70) of the maximum values of CGMS appeared during 6:00-8:59 am.,only 1％(1/70) and 3％(2/70) of them presented during 0:00-2:59 am.and 21:00-23:59 pm.,and the rest were evenly distributed for the other times of the day.In the DM group,34％(10/29) of the minimum values of CGMS were found during 0:00-2:59 am.,14％(4/29) of them appeared during 9:00-11:59 am.and 15:00-17:59 pm.,45％(13/29) of the maximum values of the CGMS presented during 6:00-8:59 am.,none was found during 21:00-23:59 pm.,0:00-2:59 am.and 3:00-5:59 am.,and the rest were evenly distributed for the other times of the day.(6) 64％(45/70) of the patients in the GDM group did not require for insulin treatment,while 36％(25/70) of them did.For those patients who received insulin treatment,after CGMS,64％(16/25) of them adjusted the insulin dosage according to their blood glucose levels.In the DM group,14％(4/29) of them did not receive insulin treatment,while for the others who did(86％,25/29); 60％ (15/25) of them adjusted the insulin dosage according to their blood glucose levels after CGMS.Conclusions Both CGMS and SMBG could correctly reflect patients' blood glucose levels.It was more difficult to control the blood glucose levels in patients with type 2 DM complicated with pregnancy than the GDM patients.Compared with SMBG,CGMS could detect postprandial hyperglycemia and nocturnal hypoglycemia more effectively.

Objective To investigate the varaiation of the incidence of macrosomia and its influencing factors. Methods A population-based study of 25 944 pregnant women,who delivered in Peking University First Hospital in term birth,with singleton,between Jan. 1,2006 and Dec. 31,2013 and accepted the gestational diabetes mellitus(GDM)screening and diagnosis,was performed. The women are grouped according to the different clinical interventions at different period. Women delivered between Jan. 1, 2006 and Dec. 31,2006 was defined as Group 2006,and they were diagnosed with glucose metabolism disorder [gestational impaired glucose tolerance(GIGT)and GDM] and intervened only when meeting National diabetes data group(NDDG)criteria. Women delivered between Jan.1,2007 and Apr. 30,2011 were defined as Group post 2007,and NDDG criteria was also applied in this period. Women delivered between May. 1,2011 and Dec. 31,2013 were defined as Group post 2011,and Ministry of Health(MOH)of China was used for GDM diagnosis in this group. All pregnant women in Group post 2007 accepted the preliminary pregnancy nutrition advice and weight management. All participants met MOH criteria were diagnosed as glucose metabolism disorder in this study,in which women diagnosed and intervened in pregnancy were defined as Group diagnosis and those not being identified during pregnancy were defined as Group missed diagnosis. It was analyzed retrospectively for the incidence of macrosomia and the influencing factor. Results (1)The prevalence of macrosomia and cesarean section was decreased every year from Jan. 2006 to Dec. 2013. The incidence of macrosomia was 9.14%in 2006,reduced to 8.02%in 2007-2011 and 6.79%in 2011-2013. The incidence of cesarean section was 55.22%,reduced to 51.04%in 2007-2011 and 44.15%in 2011-2013. However,there was not remarkable change in the prevalence of small for gestational age(P>0.05).(2)Compared with Group 2006,the incidence of cesarean section was lower in Group post 2007 [51.04%(6 504/12 744)vs 55.22%(1 371/2 483)],and the difference is significantly(P<0.05). Meanwhile,the incidence of cesarean section(44.15%,4 732/10 717)and macrosomia(6.79%,728/10 717) in Group post 2011 was lower significantly than Group 2006 and Group post 2007(P<0.05).(3)The incidence of macrosomia was 7.41%(1 129/15 227)and 6.61%(1 006/15 227)respectively in Group diagnosis and Group missed diagnosis before May 2011,combined 14.02%(2 135/15 227)in total. It was increased significantly in the incidence of GDM 21.41%(2 294/10 717)after May 2011 compared with that before (P<0.05). The incidence of macrosomia was decreased significantly using MOH criteria in GDM women since 2011. It was the downtrend in the incidence of macrosomia since 2007 in non GDM women. However,there was no difference in SGA in different period.(4)In glucose metabolism disorder women, compared with Group 2006 and Group post 2007,the incidence of macrosomia and cesarean section was lower in Group post 2011,and the difference is significantly(P<0.05). However,there was no significant difference in the incidence of macrosomia and cesarean section between Group 2006 and Group post 2007, and there was no difference in SGA in the 3 groups(P>0.05). In non GDM women,the incidence of macrosomia and cesarean section was lower in Group post 2011 than Group 2006(P<0.05);meanwhile,it was the downtrend in the incidence of macrosomia in Group post 2007 compared to Group 2006,and the difference of the incidence of cesarean section was significant(P<0.05). Conclusion The prevalence of macrosomia and cesarean section might be reduced by application of suitable criteria for diagnosis of GDM and education on nutrition during pregnancy.

Objective To analyze the characteristics of pre-gestational diabetes mellitus (PGDM) diagnosed during pregnancy (missed diagnosis before pregnancy), and to evaluate the effects of diagnostic time on pregnancy outcomes. Methods A retrospective study of 746 pregnant women who were diagnosed PGDM and delivered in Peking University First Hospital from January 1st, 2005 to December 31st, 2015 was conducted. The patients were divided into 2 group. Those diagnosed PGDM before pregnancy were defined as Group diagnosed before pregnancy, and those diagnosed during pregnancy were defined as Group diagnosed during pregnancy. In Group diagnosed during pregnancy, those diagnosed before 24 gestational weeks were defined as Group diagnosed during pregnancy A, and those diagnosed after 24 weeks were defined as Group diagnosed during pregnancy B. The prevalence of adverse pregnancy outcomes in each group were analyzed. Results (1) Rate of missed diagnosis for PGDM:the incidence of PGDM diagnosed before pregnancy was 32.2% (240/746), and those diagnosed during pregnancy (missed diagnosis before pregnancy) was 67.8% (506/746). (2) Blood glucose control during pregnancy: ①Group diagnosed before pregnancy and Group diagnosed during pregnancy: the highest glycosylated hemoglobin (HbA1c) in Group diagnosed before pregnancy was (6.6±1.1)%, higher than that in Group diagnosed during pregnancy [(6.3± 1.0)%, P=0.019]. However, there was no significant difference in the average HbA1c level between the 2 groups (P=0.616). The insulin needed percentage [90.8%(218/240) vs. 53.8%(272/506)] in Group diagnosed before pregnancy were higher than that in Group diagnosed during pregnancy (P<0.01).②Group diagnosed during pregnancy A and B:the highest HbA1c in Group diagnosed during pregnancy A was (6.9± 1.3)%, higher than that in Group diagnosed during pregnancy B [(6.1 ± 0.8)%, P<0.05]. And the average HbA1c in Group diagnosed during pregnancy A [(6.4±0.8)%] was also higher than that in Group diagnosed during pregnancy B [(6.0 ± 0.8)%, P<0.05]. In Group diagnosed during pregnancy B, 46.1%(187/406) used insulin, lower than the percentage in Group diagnosed during pregnancy A (85.0%, 85/100;P<0.01). ③There were no significant differences in the highest HbA1c and the average HbA1c between Group diagnosed during pregnancy A and Group diagnosed before pregnancy (P=0.020, P=0.037). There was neither no significant difference in the percentage used insulin during pregnancy between them (P=0.128). There were significant differences in the highest HbA1c and the average HbA1c between Group diagnosed during pregnancy B and Group diagnosed before pregnancy (P<0.01, P=0.014). There was also significant difference in the percentage used insulin during pregnancy between them (P<0.01). (3) Pregnancy outcome:①Group diagnosed before pregnancy and Group diagnosed during pregnancy: the cesarean section rate [72.5% (174/240) vs. 59.7% (302/506)] in Group diagnosed before pregnancy were higher than those in Group diagnosed during pregnancy (P<0.01). However, there were no significant differences in preterm birth rate, pre-eclampsia, macrosomia percentage, percentage of neonates being hospitalized between the 2 groups (P=0.546,P=1.000,P=0.671,P=0.804). ②There was no significant difference in preterm birth rate, cesarean delivery rate, macrosomia percentage, pre-eclampsia rate, percentage of neonates being hospitalized between Group diagnosed during pregnancy A and Group diagnosed during pregnancy B (P=0.887, P=0.495, P=0.841, P=1.000, P=1.000).③There was no significant difference in preterm birth rate, cesarean delivery rate, macrosomia percentage, pre-eclampsia rate, percentage of neonates being hospitalized between Group diagnosed during pregnancy A and Group diagnosed before pregnancy (P=0.875, P=0.093, P=0.662, P=1.000, P=0.837). The cesarean delivery rate was lower in Group diagnosed during pregnancy B than that in Group diagnosed before pregnancy (P=0.001). However, there were no significant differences in preterm birth rate, macrosomia percentage, pre-eclampsia rate, percentage of neonates being hospitalized between them (P=0.530, P=0.776, P=1.000, P=0.797). Conclusions The diagnosis of PGDM is commonly missed before pregnancy. Fasting plasma glucose should be used as screening test to identify PGDM at pre-pregnancy examination or first antenatal care. Using abnormal value of 2-hour glucose after 24 gestational weeks as the only way to diagnose PGDM is not suitable.