Macrosomia prediction using different maternal and fetal parameters in women with 50 g glucose challenge test between 130 and 140 mg/dl

The identification and treatment of women with hyperglycaemia in pregnancy: an analysis of individual participant data, systematic reviews, meta-analyses and an economic evaluation

BACKGROUND

Gestational diabetes mellitus (GDM) is associated with a higher risk of important adverse outcomes. Practice varies and the best strategy for identifying and treating GDM is unclear.

AIM

To estimate the clinical effectiveness and cost-effectiveness of strategies for identifying and treating women with GDM.

METHODS

We analysed individual participant data (IPD) from birth cohorts and conducted systematic reviews to estimate the association of maternal glucose levels with adverse perinatal outcomes; GDM prevalence; maternal characteristics/risk factors for GDM; and the effectiveness and costs of treatments. The cost-effectiveness of various strategies was estimated using a decision tree model, along with a value of information analysis to assess where future research might be worthwhile. Detailed systematic searches of MEDLINE^® and MEDLINE In-Process & Other Non-Indexed Citations^®, EMBASE, Cumulative Index to Nursing and Allied Health Literature Plus, Cochrane Central Register of Controlled Trials, Cochrane Database of Systematic Reviews, Database of Abstracts of Reviews of Effects, Health Technology Assessment database, NHS Economic Evaluation Database, Maternity and Infant Care database and the Cochrane Methodology Register were undertaken from inception up to October 2014.

RESULTS

We identified 58 studies examining maternal glucose levels and outcome associations. Analyses using IPD alone and the systematic review demonstrated continuous linear associations of fasting and post-load glucose levels with adverse perinatal outcomes, with no clear threshold below which there is no increased risk. Using IPD, we estimated glucose thresholds to identify infants at high risk of being born large for gestational age or with high adiposity; for South Asian (SA) women these thresholds were fasting and post-load glucose levels of 5.2 mmol/l and 7.2 mmol/l, respectively and for white British (WB) women they were 5.4 and 7.5 mmol/l, respectively. Prevalence using IPD and published data varied from 1.2% to 24.2% (depending on criteria and population) and was consistently two to three times higher in SA women than in WB women. Lowering thresholds to identify GDM, particularly in women of SA origin, identifies more women at risk, but increases costs. Maternal characteristics did not accurately identify women with GDM; there was limited evidence that in some populations risk factors may be useful for identifying low-risk women. Dietary modification additional to routine care reduced the risk of most adverse perinatal outcomes. Metformin (Glucophage,^® Teva UK Ltd, Eastbourne, UK) and insulin were more effective than glibenclamide (Aurobindo Pharma - Milpharm Ltd, South Ruislip, Middlesex, UK). For all strategies to identify and treat GDM, the costs exceeded the health benefits. A policy of no screening/testing or treatment offered the maximum expected net monetary benefit (NMB) of £1184 at a cost-effectiveness threshold of £20,000 per quality-adjusted life-year (QALY). The NMB for the three best-performing strategies in each category (screen only, then treat; screen, test, then treat; and test all, then treat) ranged between -£1197 and -£1210. Further research to reduce uncertainty around potential longer-term benefits for the mothers and offspring, find ways of improving the accuracy of identifying women with GDM, and reduce costs of identification and treatment would be worthwhile.

LIMITATIONS

We did not have access to IPD from populations in the UK outside of England. Few observational studies reported longer-term associations, and treatment trials have generally reported only perinatal outcomes.

CONCLUSIONS

Using the national standard cost-effectiveness threshold of £20,000 per QALY it is not cost-effective to routinely identify pregnant women for treatment of hyperglycaemia. Further research to provide evidence on longer-term outcomes, and more cost-effective ways to detect and treat GDM, would be valuable.

STUDY REGISTRATION

This study is registered as PROSPERO CRD42013004608.

FUNDING

The National Institute for Health Research Health Technology Assessment programme.

Macrosomia in non-gestational diabetes pregnancy: glucose tolerance test characteristics and feto-maternal complications in tropical Asia Pacific Australia

OBJECTIVE

To look into the glucose tolerance test characteristics and determine complications in non-gestational diabetes pregnant subjects.

METHODS

From 2006 to 2009 all non-gestational diabetes mellitus (non-GDM) pregnant women who delivered macrosomia at the North Australia's Townsville Hospital were retrospectively reviewed by extracting data from clinical record. Glucose tolerance tests results were analysed in the light of an earlier diagnosis of non-GDM.

RESULTS

Ninety-one non-GDM mothers with macrosomia were studied and compared with 41 normoglycemic subjects without macrosomia. Of the subjects with non-GDM macrosomia, 45 (49.4%) had normal 50 g glucose challenge test (GCT) without further testing, another 8 (8.8%) had abnormal GCT but normal 75 g oral glucose tolerance test (OGTT). A total of 4 (4.4%) subjects had normal GCT and OGTT. Interestingly, 14 out of 16 (87.5%) subjects who were tested with OGTT owing to past history of macrosomia had normal results but delivered macrosomic babies. Only 12 subjects had both GCT and OGTT, the rest of the cohort had either of the two tests. Subjects with non-GDM macrosomia had higher frequency of neonatal hypoglycaemia 34% as compared to 10% in non-macrosomic babies (P=0.003). Other feto-maternal complications were similar in both groups.

CONCLUSIONS

No significant pattern of glucose tolerance characteristics was identified in non-GDM mothers with macrosomic babies. In spite of being normoglycemic significant neonatal hypoglycaemia was recorded in non-GDM macrosomic babies. Further prospective studies on a larger population are needed to verify our findings.