Abstract
Gestational diabetes mellitus (GDM) is a common metabolic disorder in pregnancy
and leads to serious harm to the mother and the fetus. A variety of lncRNAs play
a key role in GDM. This meta-analysis was performed to explore the potential
value of lncRNAs in GDM diagnosis. Articles correlated with lncRNA and GDM were
screened from Embase, Medline, EBSCO, PubMed, Chinese National Knowledge
Infrastructure, and WanFang databases. Summary receiver operator characteristic
(SROC) was performed to evaluate the pooled area under curve (AUC). Forest plot
was conducted to calculate the sensitivity, specificity, diagnostic likelihood
ratio (LR), diagnostic score, and diagnostic odds ratio (DOR). Deeks’ funnel
plot was utilized to evaluate the publication bias. Eleven articles containing
12 tests (1060 GDM patients and 1066 controls) were included in this
meta-analysis. AUC (0.89, 95%CI=0.86–0.92), sensitivity (0.84, 95%CI=0.80–0.87),
and specificity (0.81, 95%CI=0.77–0.85)of the SROC curve showed a high
diagnostic value of lncRNA for GDM. Positive LR (PLR 4.40, 95%CI=3.45–5.60) and
negative LR (NLR 0.20, 95%CI=0.15–0.26) results indicated that the diagnosis of
lncRNA for GDM had low clinical utility. Diagnostic score (3.09,
95%CI=2.62–3.57) and DOR (22.04, 95%CI=13.68–35.51) results suggested lncRNAs
have good discriminative effect on GDM. Heterogeneity was significantly higher,
but not induced by the subgroups. LncRNAs had high diagnostic value and good
discriminative effect for GDM, but the clinical utility was not high. This
meta-analysis study offers a potential target for GDM diagnosis.
Keywords
gestational diabetes mellitus - diagnosis - SROC
