Gestational diabetes screening of a multiethnic, high-risk population using glycated proteins

Abstract

In populations with a high incidence of gestational diabetes (GDM), any form of oral glucose testing for screening or diagnosis excessively strains the health care system. We investigated the value of glycated proteins as potential screening tests in 430 pregnant women, i.e. protein corrected fructosamine (cFRUC) and hemoglobin A1c (HbA_1c) both alone and in combination for a GDM diagnosis confirmed by the ‘gold standard’ 100-g oral glucose tolerance test (OGTT). Two cut-off values were used for each test, the upper to rule in and the lower to rule out GDM. At the lower cut-off values for cFRUC of 210 μmol/l and HbA_1c of 5%, the sensitivities achieved were 92.2 and 92.1% while the negative predictive values were 88.9 and 86.9%, respectively. The upper cut-off values did not achieve acceptable positive predictive values to be useful for ruling in GDM. Screening of our multiethnic, high-risk pregnant population with a combination of cFRUC and HbA_1c on a single fasting sample would have avoided the cumbersome OGTT (by ruling out GDM) in 37.9% women with only a 3.9% misclassification rate. This potentially simpler approach, though not universally applicable, would be clinically useful and more acceptable to patients in selected high-risk populations.

Introduction

Gestational diabetes (GDM) or carbohydrate intolerance first recognized during pregnancy remains a controversial condition. There is no consensus on the ideal way to screen for GDM and whether screening every pregnant woman is justified. At the laboratory level, neither screening nor diagnostic testing procedures are universally agreed upon. Thus, there is a continuing uncertainty about the merits of detecting and treating this condition [1]. The recent Fourth International Workshop conference on GDM attempted to resolve many of these differences in order to achieve some degree of consensus [2]; the American Diabetes Association (ADA) expert committee on the diagnosis and classification of diabetes mellitus continues to recognize it as an important clinical diagnosis [3].

In the United Arab Emirates (UAE), previous studies have shown a very high prevalence of GDM in association with maternal obesity and high-order multiparity [4] in this significantly diverse multi-ethnic population. Despite the controversies surrounding a diagnosis of GDM, many of our physicians adhere to the screening guidelines of the American College of Obstetricians and Gynecologists (ACOG) (i.e. a non-fasting post 50-g, 1-h plasma glucose screen (GCT) followed by the confirmatory ‘gold standard’ 100-g, 3-h oral glucose tolerance test (OGTT) for positive screens and when clinically indicated). However, the recommended practice of universal screening with a GCT, particularly in high prevalence populations similar to that found locally, makes unrealistic and excessive demands on the health delivery system. These demands are further compounded by the increased reproductive rate in our patient population.

We have, therefore, continued to investigate the potential value of practical alternative screening/diagnostic tests amongst our multiracial high-risk population, which can be easily performed on a single blood sample, i.e. protein-corrected fructosamine (cFRUC) and hemoglobin A1c (HbA_1c) alone or in combination. The high prevalence of GDM in this population would be expected to increase the predictive value of these laboratory tests, while not performing an OGTT would be significantly more acceptable to patients while reducing the laboratory workload. A subsequent paper will document the performance of these tests in predicting clinical outcome as opposed to biochemical diagnostic status, and attempt to validate the results in a larger series of patients.