An Appraisal of the introduction of Metformin to the management of Gestational Diabetes Mellitus
Dr Karl McPherson, Dr Devannas Rajeswari
Ashford & St Peter’s NHS trust, Chertsey, Surrey.
Gestational diabetes mellitus (GDM) is thought to affect between 2-14% of pregnancies in the UK. GDM is often an asymptomatic disease in pregnancy that can have important short and long term effects on maternal and fetal health.
Studies have proven that increased antenatal surveillance of these patients reduces these risks and moreover, achieving euglycaemia as part of this surveillance. This can be achieved by strategic dietary modification, the use of insulin and more recently metformin.
Our aim is to elicit from a single centre experience of gestational diabetes, the efficacy and safety of metformin use both as sole antihyperglycaemic agent, and in combination with insulin in achieving euglycaemia in comparison to insulin alone.
Patients were randomly sampled from 31/12/2008 to 31/7/2010 who had been diagnosed as having gestational diabetes as confirmed by 2 hour OGTT using the WHO diagnostic criteria for GDM in the index pregnancy.
A retrospective analysis of anonymised case notes was then performed. 69 cases were identified for analysis, comprising 65 singleton pregnancies and 2 DCDA twin pregnancies. Sixty-five (98.5%) of patients trialled a low glycaemic diet as control of GDM. One patient (1.5%) had no evidence of commencing diet-control
Twenty-nine (43%) women required either Metformin or Insulin to control blood sugar, of which one was a twin pregnancy. Metformin was used in fifteen (23%) cases, of which nine (13.8%) required additional insulin.
Simple linear regression revealed where antihyperglycaemic agents were required (either Metformin and/or Insulin), the increase in maternal weight gain in pregnancy was positively correlated to increases in maternal HbA1C (p0.01).
Primiparous women were more likely to have a child affected by hypoglycaemia. This incidence appears to be independent of whether an antihyperglycaemic agent was used or not. Unsurprisingly, women with a booking BMI of greater than 30 were more likely to use antihyperglycaemic agents (P 0.049). Birth weight in this study was significantly positively correlated with maternal age, maternal booking BMI, percentage weight gain in pregnancy, HbA1c rises and interval from diagnosis of GDM to delivery. This data also suggests that maternal weight gain in pregnancy is associated with poorer neonatal 5 minute Apgar scores, although this failed to reach significance (p 0.1).
Our study was underpowered to identify a difference in rare complications such as the incidence of fetal congenital abnormalities, diabetic sequelae, Low 5 minute Apgar, fetal macrosomia and stillbirth. However observationally it is likely to be significant that these were evidenced in a cohort of just 67 patients, suggesting a higher incidence than the non-diabetic pregnant population.