Academic Editor:Giribabu Nelli, Senior Lecturer, Department of Physiology, Faculty of Medicine, University of Malaya
Checked for plagiarism: Yes
Review by: Single-blind
Pregnancy Outcome in Gestational Diabetes Mellitus under Treatment-Bangladesh Perspective
Pregnant mothers (N=191) diagnosed with GDM (n=91, age: 27.44±4.91yr; body mass index, BMI: 26.88±4.16 kg/m2; mean±SD) on the basis of WHO 2013 criteria were compared with non-GDM (n=100, age: 26.01±4.81yr, BMI: 25.53±3.77 kg/m2, mean±SD) for pregnancy outcome irrespective of gestational age. HbA1c was also measured in all mothers. Gestational hypertension, preeclampsia, premature rupture of membrane (PROM), hydramnios, recurrent urinary tract infection (UTI), recurrent moniliasis, intrauterine growth retardation (IUGR), intra uterine death (IUD), mode of delivery, birth weight, birth injury, neonatal hypoglycemia, hyperbilirubinemia, respiratory distress syndrome (RDS), congenital anomaly were recorded at every trimester. 160 mothers (GDM=75, non-GDM=85) could be followed for outcomes to the end of pregnancy. All the GDM mothers were offered standard treatment throughout pregnancy period.
HbA1c was significantly higher in GDM than that in non-GDM (5.42±0.61 vs. 4.98±0.44%, mean±SD; p<0.001).Outcome events in GDM and non-GDM were: gestational hypertension- 3.6% vs. 2.3% (p=0.621), preeclampsia- 2.4% vs. 0% (p=0.150), PROM- 4.9% vs. 0% (p=0.037), hydramnios- none in any group, recurrent UTI- 12.3% vs. 4.7% (p=0.073), recurrent moniliasis- 0.0% vs. 2.3% (p=0.165), caesarian section- 85.3% vs. 72.9% (p=0.056), small for gestational age (SGA)- 26.4% vs. 36.7% (p=0.246), large for gestational age (LGA)- 1.4% vs. 0%, p=0.246, IUGR- 2.3 vs. 2.5% (p=0.952), neonatal hypoglycemia- 2.7% vs. 0.0% (p=0.130), hyperbilirubinemia- 12.0% vs. 11.8% (p=0.963), RDS- 0.0% vs. 2.4% (p=0.181) and birth injury- 0.0% vs. 1.2% (p=0.346), congenital anomaly- 4.0% vs. 1.2% (p=0.254) and abortion- 1.3% vs. 0.0% (p=0.286). Preterm delivery (12.0% vs. 7.1%, p=0.285) and caesarean section (85.3% vs. 72.9%, p=0.056) were more in GDM.
Gestational diabetes mellitus (GDM) is increasing among the South Asians1, 2. Recently our ‘GDM study group’ of Department of Endocrinology, Bangabandhu Sheikh Mujib Medical University (BSMMU), Dhaka has observed an alarming frequency of 36.6% GDM by WHO 1999 criteria1 and 30.0% by WHO 2013 criteria along with 5.3% diabetes in pregnancy (DIP) in another study2.
GDM is associated with adverse pregnancy outcomes. However, whether standard treatment at outset makes any difference is not well studied. Several studies world wide including the land mark Hyperglycemia and Adverse Pregnancy Outcome study (HAPO study) did not assess the adverse outcomes in light of this point3. As GDM is associated with significant metabolic alterations, increased maternal and perinatal morbidity and mortality, it is imperative to screen pregnant mothers for GDM4. Many studies have observed various complications of GDM without regard to the emphasis of treatment during pregnancy 5, 6, 7, 8. A multicenter, randomized trial observed statistically significant decrease in the relative risks of several outcomes like macrosomia, large for gestational age (LGA) and shoulder dystocia with standard treatment for GDM. Additionally the risks for perinatal mortality, neonatal intensive care admission and birth trauma were also reduced in treated women9.
Present study was carried out to compare adverse pregnancy outcomes between GDM mothers on treatment and mother with normal glycemic status.
It was a prospective cohort study carried out during January, 2014 to August, 2015 by GDM study group of Dept. of Endocrinology, BSMMU. Pregnant mothers (N=191) recruited consecutively from GDM clinic and diagnosed with GDM (n=91, age: 27.44±4.91yr; body mass index, BMI: 26.88±4.16 kg/m2, mean±SD) on the basis of WHO 2013 criteria were compared with non-GDM (n=100, age: 26.01±4.81yr, BMI: 25.53±3.77 kg/m2, mean±SD) for pregnancy outcome irrespective of gestational age. Informed written consent was taken from each of the mothers.
At the time of screening oral glucose tolerance test (OGTT) as well as HbA1c were done in all recruited mothers. The GDM group received standard treatment and both the GDM and non-GDM groups were followed till delivery. Their pregnancy outcomes were recorded. Demographic, anthropometric measures, glycaemic status as well as adverse pregnancy outcomes e.g. hydramnios, recurrent monilial infections, recurrent urinary tract infection (UTI), preterm delivery, preeclampsia, macrosomia, birth injury etc. in light of the experience with GDM by our obstetricians were recorded. GDM and non-GDM mothers were advised for antenatal check-up at the end of every trimester which included clinical and biochemical parameters and ultrasound imaging for pregnancy profile. Majority of the GDM mothers were controlled with diet and exercise but some required insulin. All the mothers were advised for hospital delivery. Total 30 mothers (GDM=16 and non- GDM=14) were lost to follow-up. Prior to commencement of this study the research protocol was approved by Institutional Review Board (IRB).
Plasma glucose was assayed by RA-50 analyzer (Dade Behring, Germany). A fixed known concentration for low level (5.21 mmol/l) as well as high level (16.1 mmol/l) was used in every assay run. Inter-assay Co-efficient Variance (CV) for low level was 5.78%, and for high level was 5.59%. HbA1c was measured by the National Glycohemoglobin Standardization Program (NGSP) certified Bio-Rad D-10TM HbA1c Program 220-0101, USA.
All data were analyzed by SPSS program (version 22.0). Data were expressed either as mean (+SD/SEM) or as percentage as applicable. Comparison of pregnancy outcome variables between GDM and non-GDM were done by Chi-square test. P values ≤0.05 were considered statistically significant.
There was significant difference between GDM and non-GDM for age (27.44±4.91 vs. 26.01±4.81 yrs, mean±SD; p=0.044), BMI (26.88±4.16 vs. 25.53±3.77 kg/m2, mean±SD; p=0.020) and family history of DM (45.1% vs. 28.0%, p=0.014) (Table-I).
Complications like preeclampsia (2.4% vs. 0%, p=0.150), gestational hypertension (3.6% vs. 2.3%, p=0.621), spontaneous abortion (1.3% vs. 0.0%, p=0.286), recurrent UTI (12.3% vs. 4.7%, p=0.073), PROM (4.9% vs. 0.0%, p=0.037), preterm delivery (12.0% vs. 7.1%; p=0.285) were relatively higher in the GDM mothers while recurrent monilial infection (0.0% vs. 2.3%, p=0.165), intra-uterine growth retardation (IUGR, 2.3% vs. 2.51%, p=0.952), intrauterine death (IUD, 0.0% vs. 2.3%, p=0.167) were relatively higher in the non-GDM group (Table-II). Of the complications in infants, birth weight (2.91±0.51 vs. 2.80±0.44 kg, mean±SD; p=0.569), neonatal hypoglycemia (2.7% vs. 0.0, p=0.130), hyperbilirubininemia (12.0% vs. 11.8%, p=0.963) and congenital anomalies (4.0% vs. 1.2%, p=0.254) were relatively higher in the GDM mothers. On the other hand respiratory distress syndrome (RDS, 0.0 vs. 2.4%, p=0.181) and birth injury (0.0% vs. 1.2%, p=0.346) were relatively higher in non-GDM mothers (Table-III). On the basis of gestational age, 1.4% babies of GDM mothers were large for gestational age (LGA), but none among non-GDM. Conversely, small for gestational age (SGA) baby were 36.7% in non-GDM mothers while 26.4% in GDM mothers: (p=0.246) (Figure 1). Caesarean section was relatively more in GDM (85.3%, 64/75 vs. 72.9%, 62/85; GDM vs. non-GDM; p=0.056) which is statistically significant (Figure 2). More than 90% mothers were only on lifestyle modifications while 9% mothers required insulin for the control of glucose. As revealed by interrogation compliance to treatment by the GDM mothers. 66% mothers were found to be nicely compliant to the advices while 28% were poorly complained and about 5% were non compliant to the advices for control of glucose.Table-I. Baseline characteristics of the studied mothers
|Variables||All subjects||GDM||Non GDM|
|Age in years (mean ± SD)||26.69±4.90||27.44±4.91||26.01±4.81|
|BMI in kg/m2 (mean ± SD)||26.18±4.01||26.88±4.16||25.53±3.77|
|Family history of DM||69 (36.1)||41 (45.1)||28 (28.0)|
|Housewife||138 (72.3)||60 (65.9)||78 (78.0)|
|Service||22 (11.5)||16 (17.6)||06 (6.0)|
|Medical professional||22 (11.5)||11 (12.1)||11 (11.0)|
|Student||09 (4.7)||04 (4.4)||05 (5.0)|
|Nulliparous||89 (46.6)||43 (47.3)||46 (46.0)|
|Multiparous||102 (53.4)||48 (52.7)||54 (54.0)|
|History of abortion|
|Yes||57 (29.8)||24 (26.4)||33 (33.0)|
|No||134 (70.2)||67 (73.6)||67 (67.0)|
|SBP (mean ± SD) in mm Hg||103.51±10.50||104.62±11.01||102.50±50|
|DBP(mean ± SD) in mm Hg||65.97±9.55||66.76±10.25||65.25±8.85|
|Complications||GDM (n=75)||Non GDM (n=85)||Total (n=160)||p|
|Preeclampsia||2 (2.4)||0||2 (1.2)||0.15|
|GTN||3 (3.6)||2 (2.3)||5 (3.0)||0.621|
|Spontaneous abortion||1 (1.3)||0||1 (0.6)||0.286|
|Recurrent monilial infection||0 (0.0)||2 (2.3)||2 (1.2)||0.165|
|Recurrent UTI||10 (12.3)||4 (4.7)||14 (8.4)||0.073|
|PROM||4 (4.9)||0 (0.0)||4 (2.4)||0.037|
|IUGR||2 (2.3)||2 (2.5)||4 (2.4)||0.952|
|IUD||0||2 (2.3)||2 (1.2)||0.167|
|Complications||GDM (n=75)||Non GDM (n=85)||p|
|Birth weight (mean ± SD) in Kg||2.91±0.51||2.86±0.44||0.569|
|Neonatal Hypoglycemia||2 (2.7)||0||0.13|
|Hyperbilirubinemia||9 (12.0)||10 (11.8)||0.963|
|Congenital anomalies||3 (4.0)||1 (1.2)||0.254|
|Birth injury||0||1 (1.2)||0.346|
Our study was intended to compare the adverse outcomes in mother and infants between GDM on standard treatment with non-GDM mothers.
The study revealed that a number of maternal as well as complications with infants in GDM are higher than non-GDM. The frequency of preeclampsia and gestational hypertension were higher though not statistically significant. Similar findings were also observed by some authors though treatment was not offered in their studies8,10. In GDM group, 4 mothers developed premature rupture of membrane (PROM) which was statistically significant though under treatment. While few studies observed statistically significant higher frequency of PROM in GDM mothers without standard treatment, others did not4,10. Another study in India showed significant number of PROM among GDM mothers on treatment11. Spontaneous abortion was found in only one mother of GDM group which was similar to other studies4,12. IUD occurred in two mothers among non-GDM group which was probably due to home delivery without standard obstetric management. Recurrent UTI was more frequently found in GDM group though it was not statistically significant and seems similar to findings of another study4. Mothers with GDM are nine times more likely to have vaginal candidiasis as found in Ugandal mothers13. But interestingly our study found higher incidence of vaginal candidiasis in non- GDM mothers.
In the context of fetal outcomes neonatal hypoglycemia, hyperbilirubinemia, congenital anomalies, macrosomia were studied between two groups. The HAPO study observed continuous linear associations between hyperglycemia and hyperbilirubinemia. They found stronger association between maternal hyperglycemia with hyperbilirubinemia but weaker with neonatal hypoglycemia though HAPO included those mothers with minimal hyperglycemia without standard treatment3. Our study observed increased frequency of both neonatal hypoglycemia and hyperbilirubinemia in GDM mothers though statistically insignificant. Apropos with this, others also found no significant difference of neonatal hypoglycemia and hyperbilirubinemia between GDM and non- GDM groups14,15. On the contrary, a study in India found significantly high frequency of neonatal hypoglycemia as well as hyperbilirubinemia16. The congenital malformations among offspring of women with diabetes are not uncommon17. In our study, 4% neonates developed congenital anomalies which were not statistically different between two groups. Similar findings were observed in several studies conducted in Pakistan, India and Saudi Arabia4,10,18. This study demonstrated that only 1.4% babies are LGA which was not statistically significant3. In another study high fasting plasma glucose (FPG) was strongly associated with LGA neonates and/or macrosomia, and primary cesarean section19. Conversely, an outcome study in Jammu region showed birth weight was not significantly higher among infants of GDM mothers with treatment10.
Around 12% of the deliveries in the present study were preterm though statistically insignificant and corroborated to other studies4,20. GDM was positively related to the higher rate of caesarean section which was shown in HAPO study3. Other studies in Canada, Pakistan also showed similar type of evidence4,21. This study showed significant number of caesarean section among the GDM mothers. But the rationality behind the caesarean section in each case was not explored.
A multicentre randomized trial concluded that even very mild alterations in glucose tolerance can result in these adverse outcomes which can be prevented by simple but aggressive control of blood sugars22. In this regard, all the GDM mothers were treated with lifestyle intervention as well as insulin as required. 9% of our GDM mothers required insulin in addition to lifestyle intervention. Other studies showed higher percentage of insulin requirement12,21. Probably this lower rate of insulin requirement in our study is due to excluding mothers with DIP. It is important to mention that GDM mothers were provided with standard treatment and overall adherence to treatment was good. This might be the reason behind no significant difference of outcomes between two groups. While conducting this study on our pregnant population we confronted multiple obstacles in maintenance of good adherence to regular follow-up. Moreover, periodic follow-up for the status control by life style and insulin (where applicable) could not be meticulously maintained owing to lack of compliance in some cases which might have influence over the outcome to some extent. This also hindered the assessment over correlation between glycemic status, compliance for treatment and pregnancy outcome.
This study showed that frequency of adverse events did not differ significantly between GDM mothers and mothers with normal glycemic status. But most of the adverse events were relatively higher in GDM mothers despite standard treatment.
The authors gratefully acknowledge Novo Nordisk Pharma Bangladesh for their financial help. Departments of Biochemistry as well as Obstetrics and Gynecology of BSMMU are acknowledged for their technical helps.