Diagnostic performance of PAPP-A and β-hCG in early detection of gestational diabetes mellitus: a meta-analysis.

Abstract

Background: Gestational diabetes mellitus (GDM) is a common metabolic disorder that creates considerable risks regarding both maternal and fetal health. Conventional screening approaches for GDM which are typically performed in the late second trimester; frequently miss an important window for early intervention.

Methods: This meta-analysis seeks to evaluate the diagnostic accuracy of first-trimester maternal serum biomarkers, particularly pregnancy-associated plasma protein-A (PAPP-A) and beta-human chorionic gonadotropin (β-hCG), in the prediction of GDM. This systematic review of observational studies was conducted to assess PAPP-A and/or β-hCG levels during the first trimester, examining their correlation with the subsequent diagnosis of GDM. This meta-analysis collected data from numerous studies to evaluate sensitivity, specificity, likelihood ratios and diagnostic odds ratios; alongside with developing summary receiver operating characteristic (sROC) curves.

Results: This diagnostic meta-analysis assessed 23 studies encompassing first-trimester PAPP-A and β-hCG measurements with regards to early prediction of GDM. The overall pooled sensitivity and specificity were found to be 63% (95% CI: 53-73%) and 70% (95% CI: 61-78%), alongside an AUC of 0.72 (95% CI: 0.68-0.76). Substantial heterogeneity was observed regarding both sensitivity and specificity (I² >95%). Unaccompanied PAPP-A showed a sensitivity of 67% (95% CI: 55-77%) and specificity of 66% (95% CI: 54-76%) with AUC of 0.71, while β-hCG alone exhibited low sensitivity of 29% (95% CI: 7-69%) despite a high specificity of 87% (95% CI: 64-96%) with its AUC found to be 0.71. Fagan's analysis revealed modest clinical impact; which was found to be raising post-test probability from 20% to ~ 39% after a positive result. Deek's tests suggested no major publication bias (p = 0.45 for overall, 0.41 for PAPP-A, 0.08 for β-hCG). Subgroup analyses revealed higher sensitivity levels in studies utilizing ADA criteria and in studies with smaller samples, while those with cohort designs generated more conservative estimates upon comparison with their case-control counterparts.

Conclusion: First-trimester PAPP-A and β-hCG are found to express modest diagnostic accuracy and therefore are best considered as adjuncts to early risk stratification regarding GDM. PAPP-A, as stand-alone, provides balanced though moderate levels of sensitivity and specificity, whereas β-hCG shows high specificity levels but very low sensitivity level; thus, limiting its independent predictive value. Neither biomarker is found to be sufficient as a stand-alone diagnostic tool, but both may contribute to comprehensive risk models which might inform timely intervention. Future research should emphasize standardized methodologies and validation in large, diverse populations in order to improve clinical applicability.