Assessment of Pregnancy outcome in gestational diabetes mellitus

Abstract

Methodology for metanalysis

Cohort studies and control arms from trials detailing pregnancy complications in women with gestational diabetes mellitus met the criteria for inclusion. Stratifying studies by insulin usage, three distinct subcategories emerged: non-insulin use (patients abstaining from insulin throughout the ailment's course), insulin use (various patient segments receiving insulin treatment), and unreported insulin usage. Subsequent analyses were conducted within subgroups based on country status (developed or developing), study quality, diagnostic criteria, and screening methodology. Meta-regression models were employed, tethered to the percentage of patients undergoing insulin treatment.

Result:

In our latest investigation, we delved into 187 studies, embracing a vast cohort of 8,204,395 pregnancies. Notably, 62 studies (33.2%) exhibited a low or medium risk of bias. When insulin remained absent from the equation, adjustments for confounding factors illuminated an increased likelihood of various outcomes in women contending with gestational diabetes mellitus.

These outcomes encompassed a heightened probability of caesarean section (odds ratio 1.18, 95% confidence interval 1.05 to 1.34), preterm delivery (1.54, 1.29 to 1.85), low one-minute Apgar score (1.47, 1.04 to 2.09), macrosomia (1.75, 1.27 to 2.43), and infants born large for gestational age (1.63, 1.30 to 2.05).

Conversely, within studies incorporating insulin use and subsequent adjustments for confounding variables, women grappling with gestational diabetes mellitus exhibited elevated odds of delivering infants large for gestational age (odds ratio 1.67, 1.13 to 2.46), encountering respiratory distress syndrome (1.61, 1.22 to 2.13), neonatal jaundice (1.32, 1.05 to 1.66), or necessitating admission to the neonatal intensive care unit (2.42, 1.68 to 3.50). Our investigation revealed no conclusive evidence pointing to disparities in the odds of instrumental delivery, shoulder dystocia, postpartum haemorrhage, stillbirth, neonatal death, low five-minute Apgar score, low birth weight, and small for gestational age between women with and without gestational diabetes mellitus after adjusting for confounding variables.

Additionally, significant heterogeneity surfaced between studies concerning various adverse pregnancy outcomes, with contributing factors identified in country status, adjustment for body mass index, and screening methods.

Conclusion:

Upon meticulous adjustment for confounding variables, gestational diabetes mellitus surfaced as significantly linked to a spectrum of pregnancy complications. These revelations augment our grasp of the intricate web of adverse outcomes tethered to gestational diabetes mellitus during pregnancy. It is imperative that forthcoming primary studies systematically incorporate adjustments for a more exhaustive array of prognostic factors to enhance the robustness and applicability of their findings.