Risk stratification in early pregnancy for women at increased risk of gestational diabetes
Introduction
In the setting of increasing gestational diabetes (GDM) prevalence, adoption of the International Association of Diabetes in Pregnancy Study Group (IADPSG) diagnostic criteria is pending, with endorsement from several key international bodies, including the American Diabetes Association (ADA) [1]. Along with rising maternal obesity rates and advancing maternal age, adoption of IADPSG criteria is anticipated to increase GDM prevalence ∼2 fold to ∼15 to 20% of pregnancies [2], increasing service burden and associated costs, with many health care providers likely under-resourced to provide adequate multi-disciplinary GDM care [3]. For these reasons, disparity remains between endorsement and implementation from a clinical practice perspective.
Whilst GDM diagnosis and treatment improves clinical outcomes for mothers and babies and is cost effective [4], [5], little effort has focused earlier in pregnancy on those women at high risk of GDM, despite clinical rationale for early risk identification. Risk identification and stratification is common in maternity care, enabling streamlined services and improved clinical outcomes, as is seen with multidisciplinary care in GDM. Continuity of pregnancy care improves obstetric outcomes [6], [7], yet women diagnosed with GDM are often transferred late in pregnancy to high risk services, disrupting continuity and leaving limited opportunity for preventive intervention. Early lifestyle intervention in pregnancy limits gestational weight gain (GWG), assists in managing obesity and may also prevent GDM, reducing prevalence by 40–80% [8], [9], [10], [11]. Longer term lifestyle intervention also reduces Type 2 diabetes (Type 2 DM) [12]. Therefore, early detection of women at risk of GDM may facilitate (i) early streamlined antenatal care, (ii) enhanced continuity of care, (iii) targeted lifestyle interventions to reduce GWG, and potentially reduce GDM and Type 2 DM, (iv) timely universal GDM screening and prompt GDM management and (v) improved patient experience and clinical outcomes in the short and long term. With rising GDM prevalence and opportunities for potential prevention of GDM and its complications, there is rationale for early pregnancy GDM risk screening to identify those at high GDM risk.
Three GDM risk prediction tools have been published internationally, although these tools have not been applied using the IADPSG GDM diagnostic criteria (summarised in [13]). Our Monash GDM risk tool was developed and validated based on Australian data [14]. It was then internationally validated [13] and has also been successfully applied in a clinical setting for a targeted lifestyle randomised controlled trial (RCT) intervention [11]. The Monash tool, based on a score generated from simple GDM risk factors performs well, but could be improved further to increase clinical utility [14]. Additionally, while the role of biochemical markers in GDM risk predication have been studied with some success [15], [16], [17], clinical application is limited by cost and patient inconvenience [18]. Given the rationale for early identification of those at high GDM risk, here we undertook a novel study aiming to improve risk prediction by testing a two-step approach involving combining a validated epidemiological risk prediction tool and in those at risk only, the addition of simple biochemical markers to improve prediction of GDM risk on IADPSG diagnostic criteria.
Section snippets
Prior GDM risk prediction tool development and validation
As previously described, we undertook a retrospective cohort study at a tertiary hospital in Victoria, Australia in 2008. Routine pregnancy data in 4276 women was analysed to identify clinical risk factors for GDM [14]. Women were divided in to a derivation group (n = 2880) and a validation group (n = 1396). Derivation group data was used to develop a simple risk predictor scoring tool, with scores derived from rounded odds ratios of clinical risk factors with a possible score range of zero to
Enhancing risk prediction—Current study
In this current analysis, we explore the impact of the addition of biochemical measures to the Monash risk prediction tool in the context of the HeLP-her HRP study. Specifically, we focus on the impact of a fasting blood glucose and lipid measurement taken in early pregnancy on GDM risk prediction, in women already identified at high risk based on the Monash validated GDM risk prediction tool.
Measures
All outcome measures were completed at baseline (12–15 weeks) and 26–28 weeks gestation. Basic
Results
Of the 238 women that were recruited into the RCT and completed baseline data collection, 224 women were included in final data analysis here (Fig. 2). Of the 224, 51 (23%) were diagnosed with GDM based on the ADIPS criteria, confirming a high-risk cohort. Applying the IADPSG criteria on the 202 women that completed an OGTT would have resulted in 60 women diagnosed with GDM (30%). Baseline demographics (age, BMI, fat mass [kg], country of birth, previous pregnancies) between these subgroups
Discussion
This study builds on the application of the simple, internationally validated Monash GDM risk prediction tool, applied in early pregnancy, to identify women at high risk of GDM in a clinical pregnancy care setting. We demonstrate that the addition of early pregnancy fasting glucose, found to be the strongest predictor of GDM, strengthened the Monash GDM risk prediction tool yielding an 80% and 83% probability that we can identify women early in pregnancy who will develop GDM, on ADIPS and
Disclosure of interests
The authors have nothing to disclose. No financial disclosures were reported by the authors of this paper.
Ethics approval
The Southern Health Research Advisory and Ethics Committee approved the study and all participants gave written informed consent (Southern Health known as Monash Health as of 2013). Approval date 1/4/2008; project number 07216C.
Funding sources
This project is supported by a BRIDGES grant from the International Diabetes Federation. BRIDGES, an International Diabetes Federation project is supported by an educational grant from Lilly Diabetes (Project Number: LT07-121). The Jack Brockhoff Foundation also provided funding for this study.
Conflict of interest statement
The authors declare they have no competing interests.
Clinical trial registration
Australian New Zealand Clinical Trial Registry Number: ACTRN12608000233325. Registered 7/5/2008. [www.anzctr.org.au].
Acknowledgements
The authors would like to acknowledge Carolyn Allan, Eldho Paul and Wan Teh for input in to the original Monash GDM risk prediction tool development. In addition, we acknowledge Boyd Strauss for body composition provision, Amanda Hulley, Lauren Snell and Melanie Gibson-Helm for recruitment and data collation, Deborah Thompson and Nicole Ng for data entry. Cheryce Harrison is supported by a Postdoctoral Fellowship (100168) from the National Heart Foundation. Helena Teede is an NHMRC research
References (33)
- et al.
Treating mild gestational diabetes mellitus: a cost-effectiveness analysis
Am J Obstet Gynecol
(2011) - et al.
Collaboration in maternity care: a randomised controlled trial comparing community-based continuity of care with standard hospital care
BJOG
(2001) - et al.
Validation of early risk-prediction models for gestational diabetes based on clinical characteristics
Diabetes Res Clin Pract
(2014) - et al.
First trimester fasting hyperglycemia as a predictor for the development of gestational diabetes mellitus
Eur J Obstet Gynecol Reprod Biol
(2010) - et al.
Gestational weight gain by body mass index among US women delivering live births, 2004–2005: fueling future obesity
Am J Obstet Gynecol
(2009) Pregnancy: a teachable moment for weight control and obesity prevention
Am J Obstet Gynecol
(2010)Recurrent gestational diabetes: risk factors, diagnosis, management, and implications
Semin Perinatol
(2007)Position Statement American Diabetes Association
Diabetes Care
(2012)- et al.
National institutes of health consensus development conference statement: diagnosing gestational diabetes mellitus
NIH Consensus State Statements
(2013) - et al.
Recommended changes to diagnostic criteria for gestational diabetes: impact on workload
Aust N Z J Obstet Gynaecol
(2010)
Effect of treatment of gestational diabetes mellitus on pregnancy outcomes
N Engl J Med
Midwife-led continuity models versus othermodels of care for childbearing women
Cochrane Database Syst Rev
Lifestyle interventions for overweight and obese pregnant women to improve pregnancy outcome: systematic review and meta-analysis
BMC Med
A randomised trial of a four-step multidisciplinary approach to the antenatal care of obese pregnant women
Aust N Z J Obstet Gynaecol
Effects of interventions in pregnancy on maternal weight and obstetric outcomes: meta-analysis of randomised evidence
BMJ
Optimizing healthy gestational weight gain in women at high risk of gestational diabetes: a randomized controlled trial
Obesity (Silver Spring)
Cited by (40)
Machine learning-based models for gestational diabetes mellitus prediction before 24–28 weeks of pregnancy: A review
2022, Artificial Intelligence in MedicineCitation Excerpt :Hereunder, we will comment on the best GDM predictive models of each category, with special emphasis on those with higher quality validation and predictive power. Models of this category are based on clinical [30–45] or biochemical [46–55] variables, or on a combination of both [56–100]. These strategies are summarized on Table 1.
Triceps skinfold thickness and body mass index and the risk of gestational diabetes mellitus: Evidence from a multigenerational cohort study
2022, Obesity Research and Clinical PracticeCitation Excerpt :A recent meta-analysis reported that the overall relative risk for T2DM was almost 10 times higher in women with a history of GDM than in healthy controls [6]. Furthermore, multiple studies have suggested that women with GDM also have an increased CVD risk [4,5,7,8]. Similarly, studies have consistently shown that offspring born to GDM mothers have a greater risk of fetal and neonatal complications [9,10].
The association of general obesity, central obesity and visceral body fat with the risk of gestational diabetes mellitus: Evidence from a systematic review and meta-analysis
2021, Obesity Research and Clinical PracticeCitation Excerpt :These corresponded with the increasing trends of obesity and diabetes globally [9]. GDM increases the risk of maternal hypertension disorders of pregnancy (HDP), eclampsia and preeclampsia [3]. It also increases the risk of birth trauma, the rate of operative deliveries, neonatal respiratory issues, neonatal hypoglycaemia and macrosomia [1,4,5].
Prognosis associated with initial care of increased fasting glucose in early pregnancy: A retrospective study
2021, Diabetes and MetabolismCitation Excerpt :In the International Federation of Gynaecology and Obstetrics (FIGO) report [3], as well as in Belgium [31], China, Latin America and the UK (https://pathways.nice.org.uk/pathways/diabetes-in-pregnancy), FPG values between 5.5 and 6.9 mmol/L are considered early fasting hyperglycaemia. In our present study, none of the untreated women who developed diabetes in pregnancy had FPG levels < 5.5 mmol/L. Thus, FPG values > 5.5 mmol/L are associated with more GDM after 24 WG [14,15,23,32–34]. The results of randomized clinical trials (RCTs) are invaluable for definitively supporting guidelines advocating the immediate care of women with FPG levels ≥ 5.5 mmol/L in early pregnancy.
Persistent organic pollutants and maternal glycemic outcomes in a diverse pregnancy cohort of overweight women
2021, Environmental ResearchCitation Excerpt :Still, these measures may be informative to GDM. Studies examining early fasting glucose concentrations prior to 24 weeks gestation have found it a useful predictor of GDM risk (Smirnakis et al., 2005; Riskin-Mashiah et al. 2009; Harrison et al., 2015), and it has been proposed that a fasting glucose of ≥5.1 mmol/L before 24 weeks be used as the first pass early screening tool for dysglycemia during pregnancy (Cosson et al., 2017). Our summary measures by chemical class allowed us to examine the class-specific burden regardless of the contribution of each individual chemical; however, we do note that this method may be driven by chemicals with higher absolute concentrations.