Impact of tighter glycaemic targets on pregnancy outcomes in a multi-ethnic cohort with gestational diabetes: A retrospective,single centre observational study

Abstract

Aims

To determine the impact of tighter glycaemic targets for gestational diabetes.

Methods

Retrospective data analysis before and after introducing tighter glucose targets.

Results

In 2265 pregnancies there was no change in large-for-gestational age babies (Odds ratio (OR) 1.1; 95% Confidence interval (CI) 0.8–1.6) and macrosomia (OR 1.0; 95% CI 0.6–1.7) following tighter glucose targets. Gestational hypertension (OR 0.4; 95% CI 0.2–0.7), spontaneous vaginal birth (OR 0.6; 95% CI 0.4–0.9) reduced and pharmacological treatment increased (OR 2.2; 95% CI 1.7–2.8).

Composite adverse neonatal outcome (perinatal death, shoulder dystocia, fracture, nerve palsy) (OR 0.6; CI 0.02–14.2 p = 0.76)), Apgar<7 at 5 min, neonatal hypoglycaemia, respiratory distress syndrome or neonatal unit admission did not change but neonatal jaundice was reduced (OR 0.4; 95% CI 0.2–0.8).

Conclusions

Tighter glycaemic targets had no impact on large-for-gestational age babies or composite adverse neonatal outcomes. There was reduced spontaneous vaginal birth, gestational hypertension, neonatal jaundice and increased pharmacological treatment.

Keywords

Gestational diabetes glycaemic targets maternal and fetal outcomes single centre ethnically diverse population

Introduction

Gestational diabetes (GDM) has been defined as glucose intolerance with onset or first recognition in pregnancy.¹ GDM is associated with adverse pregnancy outcomes including large for gestational age (LGA) babies, macrosomia, gestational hypertension and pre-eclampsia.

Randomised trials have shown that treatment of GDM reduces adverse outcomes including LGA and pre-eclampsia,^2,3 but the studies’ screening criteria and target glucose values were variable. In the U.S. Maternal-Fetal Medicine Unit study, treatment targets were fasting glucose <5.3 mmol/l and 2 h post-prandial <6.7 mmol/l, which led to 7.6% needing insulin therapy, as targets were exceeded with lifestyle measures alone.³ In the Australian Carbohydrate Intolerance Study (ACHOIS), treatment targets were fasting glucose <5.5 mmol/l and post prandial 7.0 mmol/l and led to 20% needing insulin therapy.²

A systematic review of glycaemic targets for GDM, including a number of observational studies that were considered low quality, showed that a treatment fasting plasma glucose (FPG) < 5.0 mmol/l was associated with reduced risks of pre-eclampsia, LGA, neonatal hyperbilirubinaemia and neonatal hypoglycaemia.⁴ A Cochrane review found a single trial which did not report any difference in outcomes but did report an increase in the use of pharmacological therapy.⁵ The more recent TARGET study, which was a stepped-wedge cluster-randomised trial comparing tight and less tight glucose targets in 10 New Zealand hospitals, found no difference in the primary outcome of LGA.⁶ However, secondary outcomes showed discrepant findings, with improvement in neonatal outcomes but worsening of maternal outcomes.

We followed the ACHOIS glucose treatment targets until 2015, when we moved to adapted National Institute of Clinical Excellence (NICE), UK recommended glucose targets. We aimed to assess the impact of this change in guideline on maternal and neonatal outcomes in pregnant women with gestational diabetes.

Methods

Study design and population

Data were collected on all women with a singleton pregnancy diagnosed with GDM from 1st January 2013 to 31st December 2017 at the Royal London Hospital, a large tertiary centre in East London, serving a predominantly minority ethnic population. Clinical data were collected routinely via yearly clinical audits, registered with the Barts Health NHS Trust (UK) Clinical Effectiveness group (Audit no. 9874).

GDM screening and management

Screening for GDM was conducted using a 2-h (0 min and 120 min) 75 g oral glucose tolerance test (OGTT) after overnight fasting. This was offered between 24 and 28 weeks’ gestation to women with risk factors, such as obesity (body mass index at or above 30 kg/m²), previous macrosomic baby (4.5 kg or more), previous GDM, family history of diabetes (first degree relative with diabetes) or minority ethnic origin with a high prevalence of diabetes. Women with previous GDM were also screened at 16 weeks’ gestation using the same method.

Between 1st January 2013 and 31st August 2015, diagnostic criteria for GDM were fasting plasma glucose (FPG) ≥ 5.8 mmol/l (104.4 mg/dl) and/or 2-h ≥ 7.8 mmol/l (140.4 mg/dl). Women diagnosed with GDM received standardised dietary education available in English and Bengali languages. Those not achieving the glycaemic targets of fasting <5.5 mmol/l (99 mg/dl) and/or 2 h post-prandial <7.0 mmol/l (126 mg/dl) were started on metformin and/or insulin. These screening values and glycaemic targets had been in place for several years with good clinical outcomes in regular audits. From September 1st 2015 to 31st December 2017, a new clinical guideline was introduced. This contained three key changes in clinical practice. Firstly, GDM was diagnosed with one of FBG ≥ 5.6 mmol/l (100.8 mg/dl) or 2-h value ≥ 7.8 mmol/l (140.4 mg/dl). Secondly, glycated haemoglobin (HbA1c) was added to booking bloods for all women with previous GDM, diabetes in a first degree relative, BMI ≥ 30 kg/m2 or a previous baby weighing 4 kg or more at birth.

An HbA1c between 41 and 47 mmol/mol (5.9–6.4%) triggered an early OGTT and five days of capillary blood glucose (CBG) monitoring with results reviewed in a multi-disciplinary clinic to ascertain whether a diagnosis of GDM could be made. Thirdly, the glycaemic control targets were amended to FBG < 5.5 mmol/l (99 mg/dl) or 1-h post-prandial value <7.8 mmol/l (140.4 mg/dl). The FBG<5.5 mmol/l (99 mg/dl) was maintained to minimise the expected pressure on clinical resources with the new guideline. Routine fetal scans took place at 12 weeks and 20 weeks for both groups. Additional growth scans in the treatment group took place at 28, 32, 36 and 38 weeks’ gestation. Women needing lifestyle measures alone could deliver spontaneously up to 40 + 6 weeks, whereas women needing pharmacological therapy were advised to deliver between 38 and 39 + 6 weeks.

Throughout this time period, there was no significant change in the clinical leadership for diabetes (MSBH) or obstetrics (AS, ST), and no significant variation in obstetric practice.

Process

Data were collected retrospectively from electronic health records and cleaned to exclude duplicates. Women with pre-existing diabetes were excluded. Baseline maternal characteristics including age, ethnicity, body mass index (BMI) at booking, parity, and gestational age (as calculated by dating ultrasound or LMP) as well as outcome data were extracted from the electronic hospital records.

Outcomes

Clinically relevant outcomes were decided by clinician authors a priori, informed by previous GDM reports.^3,6,7 Maternal outcomes: induction of labour (IOL), mode of delivery, pre-term birth (birth before 37 weeks’ gestation and gestational hypertension (sustained new systolic blood pressure ≥ 140 mmHg or diastolic blood pressure ≥90 mmHg after 20 weeks' gestation), pre-eclampsia (hypertension with urine protein-creatinine ratio (PCR) > 0.03 g/mmol or other clinical features of pre eclampsia). Shoulder dystocia (impacted anterior shoulder preventing pelvic descent) was reported for vaginal deliveries.

Neonatal outcomes: LGA (defined as WHO birthweight ≥90th centile), macrosomia (defined as birthweight ≥4.5 kg and ≥ 4.0 kg in women of White and minority ethnic origin, respectively), neonatal hypoglycaemia ((<2.6 mmol/l), low Apgar score (≤7 at 5 min), admission to the neonatal unit (neonatal transitional care unit (TCU), neonatal special care baby unit (SCBU) or neonatal intensive care unit (NICU)) stillbirth (any death between 24 weeks and birth), neonatal death (any death within 28 days of life) respiratory distress syndrome, neonatal jaundice requiring phototherapy.

In addition, we recorded composite perinatal outcomes, as per ACHOIS trial (perinatal death, bone fracture, shoulder dystocia).² Birthweight centiles were calculated using the World Health Organisation (WHO) WHO birthweight centile calculator.⁸

Statistical analyses

Continuous data was assessed for normal distribution (Shapiro-Wilks test), presented as mean ± standard deviation (SD) or median ± interquartile range (IQR) and were compared using Student's T test or Mann-Whitney U test as appropriate. Categorical variables were compared using Chi-squared test or Fisher's exact test. The cohort with less tight glucose targets (FPG<5.5 mmol/l and 2 h value <7.0 mmol/l)) were compared with the cohort with tighter glucose targets (FPG<5.5 mmol/l and 1 h value<7.8 mmol/l). Women who did not complete the pregnancy or completed the pregnancy in another centre were included in the analyses of maternal characteristics only. Maternal and neonatal characteristics were compared using univariate analyses. A binary logistic regression model was used adjusted for age, parity, and ethnicity. The co-variates that were entered into the model included type of treatment (lifestyle or pharmacological), type of delivery (vaginal or instrumental or caesarean), onset of labour (spontaneous or induced), type of caesarean delivery (emergency or elective), gestational hypertension, pre-eclampsia (PET), macrosomia, birthweight ≥ 90^th centile, birthweight ≤ 10^th centile, Apgar ≥7 at 5 min, neonatal hypoglycaemia, intrauterine death (IUD), neonatal death, shoulder dystocia, fracture/nerve palsy, neonatal jaundice, respiratory distress syndrome (RDS), composite neonatal outcome (including intrauterine/neonatal death, shoulder dystocia, fracture/nerve palsy), admission to neonatal unit. The statistical software package Statistical Package for the Social Sciences (SPSS) version 25 was used for analysis.

Results

Overall, 2265 women delivered between 1^st January 2013 and 31^st December 2017. This included 1177 women with GDM who delivered before guideline change with less tight glucose targets and 1088 women after guideline change with tighter glycaemic targets. Maternal characteristics are shown in Table 1. There were no differences in BMI, ethnicity or parity between the two groups, but the women with tighter glycaemic control were older at booking (32.4 ± 5.3 versus 31.8 ± 5.1 years, p = 0.006). The median fasting glucose and 2-h glucose values during oral glucose tolerance test (OGTT) did not change. The altered OGTT screening criteria (fasting glucose 5.6 mmol/l v 5.8 mmol/l) led to an extra diagnosis in 33/1025 (3.2%). There was no measurable difference between early (<20 weeks) and late (>20 weeks) GDM diagnosis.

Table 1.

Maternal baseline characteristics and details of GDM screening compared less tight and tight glucose targets.

		Pre-guideline change (n = 1173) (less tight glucose targets)	Post-guideline change (n = 1088) (tighter glucose targets)	P value
Age (years)±SD		31.8 ± 5.1	32.4 ± 5.3	0.006
BMI (kg/m²)		27.0 ± 7.9	27.5 ± 5.0	0.06
Ethnicity	White	87/1169 (7.4%)	86/1084 (7.9%)	0.28
	South Asian	937/1169 (80.1%)	882/1084 (81.3%)
	Black	93/1169 (7.9%)	72/1084 (6.6%)
	East Asian	48/1169 (4.1%)	35/1084 (3.2%)
	Middle Eastern	4/1169 (0.3%)	9/1084 (0.8%)
Parity	0	352/1168 (30.1%)	350/1117 (26.9%)	0.47
	1	322/1168 (27.5%)	300/1117 (26.9%)
	2	152/1168 (23.4%)	123/1117 (11%)
	≥3	274/1168 (23.4%)	240/1117 (21.4%)
Fasting glucose (mmol/l)		4.9 (3.9-5.9)*	4.8 (3.7-5.9)	0.19
2 h glucose (mmol/l)		8.6 (6.9-10.3)*	8.6 (7.0-10.2)	0.12
OGTT timing	<20 weeks	106/806 (13.1%)	163/988 (16.4%)	0.05
	>20 weeks	700/806 (86.8%)	825/988 (83.5%)	0.05

* Median (Interquartile range).

Maternal outcomes with tight and less tight glucose targets are shown in Table 2a. There was a significant reduction in spontaneous vaginal deliveries with tight compared to less tight glucose targets (480/165 (45.1%) versus 676/1077 (62.8%), OR 0.6 (CI 0.4–0.9) p = 0.005). There was no measurable difference in instrumental or Caesarean delivery rates. There was an increase in elective Caesarean deliveries with tight targets (149/326 (45.7%) versus 106/366 (31.6%), OR 1.8 (CI 1.3 2.7), p = 0.001). There was no change in pre-term delivery, with an overall reduction in gestational age at delivery by one day with tight targets (272 (IQR 14) days versus 273 (IQR 15) days, OR 0.98 (CI 0.97–1.0) p = 0.008). There was a reduction in birthweight <10^th centile with tight targets (105/1064 (9.8%) versus less tight targets 140/1176 (11.9%), OR 0.6 (CI 0.4–0.9), p = 0.01). There was a significant increase in women needing pharmacological treatment with tight targets (381/1079 (35.2%) versus less tight targets 172/1080 (15.9%), OR 2.2 (CI 1.7–2.8) p < 0.001). There was a significant reduction in gestational hypertension (OR 0.4 (CI 0.2–0.7 p = 0.002)) but no change in pre-eclampsia.

Table 2A.

Unadjusted and adjusted maternal outcomes comparing pre-guideline and post-guideline.

	Pre-guideline change (less tight glucose targets)	Post-guideline change (tighter glucose targets)	Unadjusted P value	Adjusted OR (95% CI)*	P-value
Primary outcome
Birthweight>90^th centile	181/1176 (15.3%)	139/1084 (12.8%)	0.08	1.1 (0.8-1.6)	0.42
Macrosomia	71/1104 (6.4%)	58/1084 (5.6%)	0.48	1.0 (0.6-1.7)	0.97
Secondary maternal outcomes
Gestational hypertension	59/1114 (5.3%)	33/1035 (3.2%)	0.01	0.4 (0.2-0.7)	0.002
Pre-eclampsia	13/1115 (1.2%)	18/1060 (1.7%)	0.31	0.4 (0.2-1.0)	0.06
Mode of labour
Induction of labour	254/1077 (23.6%)	414/1065 (38.9%)	<0.001	1.3 (0.9-1.8)	0.2
Spontaneous labour	676/1077 (62.8%)	480/1065 (45.1%)	<0.001	0.6 (0.4-0.9)	0.005
Mode of delivery
Spontaneous vaginal delivery	627/1076 (58.2%)	605/1067 (56.7%)	0.46	1.4 (1-1.9)	0.06
Instrumental delivery	131/1076 (12.2%)	135/1067 (12.7%)	0.73	1.1 (0.08-18.7)	0.06
Caesarean section	318/1076 (29.6%)	327/1067 (30.6%)	0.76	NA	NA
Elective caesarean section	106/336 (31.6%)	149/326 (45.7%)	0.01	1.8 (1.3-2.7)	0.001
Emergency caesarean section	226/336 (67.5%)	176/326 (63.8%)	<0.001	0.9 (0.6-1.1)	0.4
Pre-term birth	110/1137 (9.7%)	112/1066 (10.5%)	0.52	0.8 (0.5-1.3)	0.3
Gestational age at delivery (days)	273 (IQR 15)	272 (IQR 14)	<0.001	0.98 (0.97-1.0)	0.008
Birthweight <10^th centile	140/1176 (11.9%)	105/1064 (9.8%)	0.09	0.6 (0.4-0.9)	0.01
Birthweight (g)	3163 ± 563	3121 ± 578	0.09	1 (1-1.0)	0.4
Treatment including insulin/metformin	172/1080 (15.9%)	381/1079 (35.2%)	<0.001	2.2 (1.7-2.8)	<0.001

*Adjusted for age, BMI, parity and ethnicity.

Table 2b shows fetal and neonatal outcomes. The key outcomes of LGA and macrosomia were not measurably different. There was no change in the composite neonatal outcome with tight compared to less tight targets (13/1082 (1.2%) versus 12/1174 (1.0%), OR 0.6 (CI 0.02–14.2) p = 0.76). There were no measurable differences in Apgar<7 at 5 min, neonatal hypoglycaemia, shoulder dystocia, respiratory distress syndrome (RDS), nerve palsy/fracture, intrauterine death, or neonatal unit admission. There was a reduction in neonatal jaundice with tight targets (28/1063 (2.6%) versus 41/1175 (3.5%), OR 0.4 (CI 0.2–0.8), p = 0.01).

Table 2B.

Table showing fetal outcomes pre- and post-guideline change*.

	Pre-guideline change (less tight glucose targets)	Post-guideline change (tighter glucose targets)	Unadjusted P value	Adjusted OR (95% CI)	P-value
Composite neonatal outcome	12/1174 (1.0%)	13/1082 (1.2%)	0.7	0.6(0.02-14.2)	0.76
Apgar<7 at 5 min	98/1176 (8.3%)	194/1084 (17.8%)	<0.001	1.1 (0.7-1.8)	0.55
Neonatal hypoglycaemia	8/1175 (0.7%)	7/1063 (0.7%)	0.94	1.2 (0.2-4.0)	0.76
Shoulder dystocia	4/1175 (0.3%)	5/1063 (0.5%)	0.74	1.6 (0.1-53.4)	0.78
Jaundice	41/1175 (3.5%)	28/1063 (2.6%)	0.25	0.4 (0.2-0.8)	0.01
Respiratory distress syndrome	25/1175(2.1%)	15/1063 (1.4%)	0.20	1.8 (0.8-4.1)	0.16
Nerve palsy/fracture	0	2/1063 (0.2%)	NA	0.6 (0.02-14.2)	0.76
Intrauterine death	7/1175 (0.6%)	6/1063 (0.6%)	0.92	1.9 (0.1-64.7)	0.73
Neonatal admission†	86/1176 (7.3%)	105/1084 (9.7%%)	0.06	1.1 (0.7-1.6)	0.79

†Neonatal intensive care (NICU), Special Care Baby Unit (SCBU), Transitional Care Unit (TCU).

*Adjusted for age, BMI, parity and ethnicity.

Discussion

Our single centre, observational study showed that there were no changes in the key outcomes of LGA/macrosomia or in composite neonatal outcome following implementation of NICE screening and tighter glucose targets in our population. There were reductions in gestational hypertension, SGA babies, and neonatal jaundice. There was a measurable reduction in spontaneous vaginal delivery, increases in induction of labour and elective caesarean delivery, and a doubling of women needing pharmacological therapy.

The lack of effect on LGA or macrosomia is in keeping with the TARGET trial, which similarly showed no difference in this primary outcome.⁶ Whilst it is accepted that treatment of GDM to control maternal hyperglycaemia reduces LGA infants and other associated outcomes, it is not clear that tightening of glucose targets leads to additional benefit. Our study and the TARGET trial do not support the use of tighter glucose targets to reduce LGA infants. Our findings do provide reassurance that use of the older ACHOIS based glucose targets, produces similar outcomes on infant size as the tighter targets.²

Many fetal and neonatal outcomes are associated with LGA infants, so it follows that our study confirms no difference in the composite neonatal outcome of perinatal death, shoulder dystocia, nerve palsy and fracture. This contrasts with the TARGET trial where the composite of serious neonatal health outcome (stillbirth, neonatal death, birth trauma, shoulder dystocia) was reduced in the tighter glucose target arm. This reduction was however likely driven by three stillbirths in the less tight glucose target arm. There were no details reported in the paper as to the proposed aetiology of stillbirth in these cases; not all stillbirths are associated with glucose control and gestational diabetes.

The TARGET study was performed over a relatively short period; the effect of a small fluctuation in the rare incidences of stillbirth should be considered. Abell and colleagues examined the effects of tight and less tight glucose targets in two different services in Australia over a 4 year period.⁹ In keeping with our study, they found no difference in the primary outcome of LGA infants or the composite perinatal outcome (perinatal death, shoulder dystocia, fracture, nerve palsy) between the two centres.

Our data also found no difference in neonatal unit admission, Apgar score <7 at 5 min, respiratory distress syndrome or neonatal hypoglycaemia. This is in keeping with the TARGET trial which did not find any difference in neonatal admissions or length of stay. Our study does however report a measurable reduction in neonatal jaundice with tighter glucose targets. This was not found in the TARGET trial, but similar findings were reported by Abell and colleagues. The Hyperglycaemia in Pregnancy Outcomes (HAPO) study did report a weak association between glucose concentrations and hyperbilirubinaemia.¹⁰

Medical intervention for mothers increased with tighter glucose targets. There were less spontaneous vaginal deliveries, and although the overall rate of caesarean deliveries did not change, the proportion of elective caesarean sections increased with tighter control. Induction of labour was not measurably different when adjusted for other variables. As the number of women needing pharmacological treatment doubled, the decrease in spontaneous deliveries is not surprising. In our guidelines, women needing lifestyle measures alone can deliver spontaneously up to 40 + 6 weeks, whereas women needing pharmacological therapy are advised to deliver between 38 and 39 + 6 weeks. Moving more women into the pharmacological treatment arm therefore increased the advice to deliver earlier, and often with induction or planned caesarean section. This significantly increases medical intervention for women and can put extra pressures on hospital resources. There was an unexpected finding of reduced SGA babies with the tighter glucose targets and is difficult to explain this physiologically.

Our study was observational and the change in glucose targets was also accompanied by two other guideline changes. Firstly, the screening criteria changed slightly – fasting glucose reduced from 5.8 mmol/l to 5.6 mmol/l. However, this only resulted in additional 33 women being diagnosed with GDM so it is unlikely to have had a significant effect on the interpretation of the data. Secondly, we introduced HbA1c at booking to try and identify undiagnosed type 2 diabetes in our population, and help risk stratify at delivery. This did not have a significant effect on the number of women diagnosed with GDM less than 20 weeks, so it is unlikely that this had a significant effect on overall outcomes (the overall number of positive OGTT <20 weeks changed from 13.1 to 16.1% but this was not statistically significant (Table 1)). This is relevant as the recent TOBoGM study has shown that treatment of women at a mean of 15 weeks' gestation leads to modest benefits in neonatal outcomes, although no significant impact on neonatal adiposity.¹¹ Using Hba1c at booking to pick up women with undiagnosed type 2 diabetes will likely lead to more early OGTTs being carried out (Hba1c 41–47 mmol/mol triggered OGTT in our pathway) and it is possible that both additional lower risk women and higher risk women were picked up with this approach. However, in our analyses, as above, it did not significantly increase the number of women diagnosed <20 weeks' gestation. Based upon these observations, we have concluded that the change in glucose target was likely to have produced the most significant impact on subsequent outcomes.

This study is assessing the effects of tighter glucose targets as recommended by NICE, UK guidelines, on our local population in the UK. It is recognised however that NICE differs from other guidelines internationally in certain aspects. For instance, NICE recommends metformin as first line therapy, followed by insulin and targets of fasting <5.3 mmol/l, 1 h at 7.8 mmol/l and 2 h at 6.4 mmol/l; the American Diabetic Association (ADA) Diabetes in pregnancy guidelines advise insulin as first line therapy and glycaemic targets of fasting <5.3 mmol/l, 1 h at <7.8 mmol/l and 2 h at <6.7 mmol/l.

It is appropriate to discuss whether a glucose target of fasting 5.5 mmol/l and 1 h value of 7.8 mmol/l is actually ‘tighter’ than the glucose target of 5.5 mmol/l and 2 h value of 7.0 mmol/l. The assumption is partially based upon the expected peak of glucose after a meal in women with GDM. A small study using continuous glucose monitoring (CGM) in South Asian women showed a peak glucose at 64.3 ± 11.6 min, and other older data have suggested similar peaks in other populations.^12,13 The alternative NICE glucose target of 2 h at 6.7 mmol/l (derived from the same data and thought to be roughly equivalent to the 1 h 7.8 mmol/l value) would also be slightly ‘tighter’ than our original glucose targets. Therefore, on balance it is likely that in our predominantly South Asian population, the newer glucose targets would have been tighter than the previous ones.

Our study has strengths in that we report data for a large number of women over a five-year period in a single centre. Differences in clinical practice between centres can be very difficult to adjust for, using statistical analysis or trial design, and these can make studies comparing different centres harder to interpret. Guideline changes also take time to effect clinical practice and outcomes, with inevitable overlap between guidelines. The TARGET trial study design, analysing data from ten centres with different implementation dates/time periods of tighter targets, may have affected outcome data. Our study had no change in senior clinicians over the observed period, with no other significant population changes (such as Covid-19 pandemic) occurring during this time period.

The main limitation of the study is that there would have been a small overlap when changing guidelines in September 2015. Some women would have delivered shortly after the guideline change and would have been managed with less tight targets. It is difficult to adjust for this overlap without excluding women from the study and given the large numbers of women analysed over a long time period, the effect of the overlap is likely to be minimal. Although tighter glucose targets were mandated with the guideline change, our data does not show whether these resulted in reduced glucose levels, and this is a limitation of both our study and the recent TARGET trial. It is acknowledged that the differential effects on different treatment combinations (e.g. metformin vs insulin and metformin vs insulin alone) could not be interrogated with our current data, and it was not possible to adjust for these treatment combinations in the analyses. The recent meta-analysis of sixteen randomised controlled trials is reassuring that metformin appears to have little effect on small for gestational babies, preterm labour or caesarean section rate, but the longer term follow up is limited.¹⁴ It is also acknowledged that we did not have data on maternal hypoglycaemia, which may have affected some of the maternal outcomes.

In conclusion, we have shown that in a single centre, the effect of tightening glucose targets did not have any significant effect on LGA, macrosomia or a composite neonatal outcome in our population. However, tighter glucose targets doubled the number of women needing pharmacological treatment, reduced spontaneous vaginal deliveries and increased the proportion of elective Caesarean sections and numbers of women needing induction, impacting hospital resources.

Footnotes

Acknowledgements

The authors would like to acknowledge the following who helped collect the clinical data: Rong Lu, Sigrun Kabeya, Caroline Kinuga, Adalina Sacco, Mohammed Halder and Edson Nogueira.

ORCID iDs

Mohammed SB Huda

Nafisa Islam

Kate Wiles

Ethical considerations

As above, these data were collected as part of a registered clinical audit and does not require ethical approval.

Author contributions

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Mohammed SB Huda, Niki Margari, Nafisa Islam, Hannah Jaumdally, Kate Wiles, Rehan Khan, Peter Jacob, Elena Greco, Stamatina Iliodromiti, Anita Sanghi, Shakila Thangaratinam. The first draft of the manuscript was written by Mohammed SB Huda and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability

Source data is available for further enquiry and to share in a relevant public data repository.

Guarantor

Dr MSB Huda

Anonymity

No identifying information that would compromise anonymity.

References

Alberti

Zimmet

. Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: diagnosis and classification of diabetes mellitus provisional report of a WHO consultation. Diabet Med 1998; 15: 539–553.

Crowther

Hiller

Moss

, et al. Effect of treatment of gestational diabetes mellitus on pregnancy outcomes. N Engl J Med 2005; 352: 2477–2486.

Landon

Spong

Thom

, et al. A multicenter, randomized trial of treatment for mild gestational diabetes. N Engl J Med 2009; 361: 1339–1348.

Prutsky

Domecq

Wang

, et al. Glucose targets in pregnant women with diabetes: a systematic review and meta-analysis. J Clin Endocrinol Metab 2013; 98: 4319–4324.

Martis

Brown

Alsweiler

, et al. Different intensities of glycaemic control for women with gestational diabetes mellitus. Cochrane Database Syst Rev 2016; 4: CD011624.

Crowther

Samuel

Hughes

, et al. Tighter or less tight glycaemic targets for women with gestational diabetes mellitus for reducing maternal and perinatal morbidity: a stepped-wedge, cluster-randomised trial. PLoS Med 2022; 19: e1004087.

International Association of Diabetes in Pregnancy Study Group Working Group on Outcome

Feig

Corcoy

, et al. Diabetes in pregnancy outcomes: a systematic review and proposed codification of definitions. Diabetes Metab Res Rev 2015; 31: 680–690.

Norris

Seaton

Manktelow

, et al. Updated birth weight centiles for England and Wales. Arch Dis Child Fetal Neonatal Ed 2018; 103: F577–F582.

Abell

Boyle

Earnest

, et al. Impact of different glycaemic treatment targets on pregnancy outcomes in gestational diabetes. Diabet Med 2019; 36: 177–183.

10.

Group

HSCR

Metzger

Lowe

, et al. Hyperglycemia and adverse pregnancy outcomes. N Engl J Med 2008; 358: 1991–2002.

11.

Simmons

Immanuel

Hague

, et al. Treatment of gestational diabetes Mellitus diagnosed early in pregnancy. N Engl J Med 2023; 388: 2132–2144.

12.

Louie

Markovic

Ross

, et al. Timing of peak blood glucose after breakfast meals of different glycemic index in women with gestational diabetes. Nutrients 2012; 5: –9.

13.

Singh

Gupta

Goyal

, et al. Glycemic profile of women with normoglycemia and gestational diabetes mellitus during early pregnancy using continuous glucose monitoring system. Diabetes Res Clin Pract 2020; 169: 108409.

14.

Zhang

Ren

, et al. Metformin use in gestational diabetes mellitus and neonatal outcomes: a systematic review and meta-analysis on the risk of small for gestational age. Front Med (Lausanne) 2025; 12: 1737337.