Oral Concurrent Session 6 - Ultrasound and Genetics
70 - Universal fetal echocardiography for pre-gestational diabetes: a cost-effectiveness analysis
Friday, January 29, 2021
5:30 PM – 5:45 PM EST
Objective: Pre-gestational diabetes mellitus (DM) is associated with an increased risk of congenital heart disease (CHD), which increases with rising hemoglobin A1c (HbA1c). Currently, all pregnant patients with DM are recommended to have a fetal echocardiogram (echo) to evaluate for CHD. Our study sought to evaluate whether there is a HbA1c level below which patients with DM should only have a fetal echo after an abnormal detailed anatomy ultrasound.
Study Design: We developed a cost-effectiveness model comparing outcomes for pregnant patients with DM with varying HbA1c levels. Probabilities and costs were derived from the literature. We used individual models to evaluate different scenarios: first trimester HbA1c < 6.5%, HbA1c 6.5-9.0%, and HbA1c > 9.0%. Primary outcomes included intrauterine fetal demise, neonatal demise, false positive, and false negative results. A cost-effectiveness threshold was set at $100,000 per quality adjusted life year (QALY).
Results: Universal echo is not cost-effective except for when first trimester HbA1c is > 9.0% (incremental cost-effectiveness ratio (ICER) of $357,367, $205,763, and $34,591 for HbA1c < 6.5%, 6.5-9.0%, and > 9.0%, respectively). The model is sensitive to changes in the probability of CHD at a given HbA1c level, as well as the cost of a neonatal transfer to a higher level of care. In Monte Carlo simulation, the probability that universal echo is cost-effective is 32.1%, 49.7%, and 64.5% for each of the three models respectively.
Conclusion: For pregnant patients with a first trimester HbA1c < 6.5%, universal fetal echo is not cost-effective, whereas for those with a first trimester HbA1c > 9.0%, universal fetal echo is indicated. For pregnant patients with a HbA1c between 6.5 and 9.0%, universal fetal echo is cost-effective in about 50% of cases, therefore clinical judgement based on individual patient factors, willingness to pay to detect CHD, and resource availability is needed. This model may help provide guidance regarding appropriate test utilization in pregnant patients with DM.