Preeclampsia: the triglyceride-glucose index remains inadequate for stand-alone screening

An association examined in twelve studies

Insulin resistance plays a role in preeclampsia. The authors therefore examined the potential value of the triglyceride-glucose index, or TyG, as a metabolic marker associated with this complication. The question concerned both whether an association exists and whether the marker can distinguish affected women from others.

Their systematic review searched eight databases for observational studies published through December 31, 2025. Twelve studies were included. The source dossier provides access to the PubMed abstract, not the full text.

A statistical signal, not a directly quantifiable increase in risk

A higher TyG index was associated with higher odds of preeclampsia, with a pooled odds ratio of 1.85 and a 95% confidence interval of 1.48 to 2.32. This result does not mean that individual risk increases by 85%. The abstract does not provide the absolute risks needed to translate this association into a concrete probability.

Results varied substantially across studies, with heterogeneity of 71.2%. The pooled estimate therefore does not describe a uniform effect across all populations and measurement conditions.

Limited discriminatory performance

In the five studies assessing diagnostic performance, pooled sensitivity was 0.65 and specificity was 0.58. The area under the curve was 0.64, with a 95% confidence interval of 0.59 to 0.68. These results indicate a limited ability to distinguish women with preeclampsia from those without it.

Crucially, some studies measured TyG close to the onset of preeclampsia, or even after it had developed. Detecting a difference when the disease is already present does not demonstrate that it can be predicted.

A potential role still to be validated

Chinese populations predominated in the included studies, limiting the generalizability of the findings. The authors call for prospective validation in diverse populations, with standardized measurement timing and predefined thresholds.

The potential value of TyG would need to be assessed alongside other variables in prediction models. This review does not demonstrate that it improves these models, or that changing this index would reduce the occurrence of preeclampsia.

What we can take away

The TyG index is associated with preeclampsia, but association and prediction remain two distinct questions. These data do not support its use alone for screening.