An Evaluation of Diagnostic Approaches for Gestational Diabetes Screening in South Africa

Mennen, Mathilda

An Evaluation of Diagnostic Approaches for Gestational Diabetes Screening in South Africa

Thesis / Dissertation

2024

Publisher

University of Cape Town

Department

Department of Public Health and Family Medicine

Faculty

Faculty of Health Sciences

Abstract

Background: Gestational diabetes mellitus (GDM) poses substantial risks to both mothers and their offspring. In South Africa, screening practices vary, and pregnant women are not screened universally due to resource constraints. This study investigates the implications of using point of-care (POC) capillary glucose measures for GDM screening and explores potential strategies to increase screening capacity by eliminating the reliance on central laboratory facilities or reducing the time spent at antenatal facilities for mothers. Objectives: The prevalence of GDM determined by venous blood glucose (VBG) measures obtained during 2-hour oral glucose tolerance tests (OGTTs) was compared to POC capillary glucose (CBG) tests with immediate results. The agreement between VBG and CBG measures was calculated across the whole cohort and in sub-groups, and the clinical and cost implications of each method explored. Methods: A secondary analysis was conducted on data collected from 400 pregnant participants who were enrolled at 24-28 weeks' gestation into a prospective cohort study at an antenatal clinic in Cape Town. Participants were screened for GDM using a gold-standard 75g OGTT and simultaneously underwent POC capillary glucose testing. GDM was diagnosed via each method according to the International Association of Diabetes and Pregnancy Study Groups (IADPSG) criteria. We calculated the sensitivity and specificity of CBG in detecting VBG-defined GDM at different thresholds and Bland-Altman analyses examined agreement between CBG and VBG. Results: The GDM prevalence was 7% among all participants, resulting from a prevalence of 6% among patients with no risk factors, and 8% among patients with risk factors. Four percent of the cohort was diagnosed with GDM despite having no risk factors. Most cases (96%) were diagnosed on fasting venous measures. Capillary measures overestimated the prevalence of GDM at IADPSG thresholds (25%) and had poor sensitivity (73%). Correlation between venous and capillary measures was lowest in the fasting state (r=0.22, p<0.001). Bland Altman analyses found the average agreement between methods to be lowest in the fasting state. 2 Conclusion: Capillary measures demonstrate poor correlation and agreement with venous measures at 24-28 weeks' gestation, particularly in the fasting state when almost all GDM cases are diagnosed. A fasting plasma glucose, if performed universally as a single measure, outperforms selective risk factor-based OGTT screening and fasting capillary blood glucose in terms of sensitivity and specificity, while reducing the overall number of laboratory-dependent glucose tests performed.

Keywords

public health

Reference:

Collections

Masters

Full item page