Does High Lp(a) Plus Insulin Resistance Signal Higher Heart Risk?

Two heart-risk markers that get a lot of attention, lipoprotein(a) [Lp(a)] and a calculated insulin-resistance score called the estimated glucose disposal rate (eGDR), were tested together in a large cohort in a study published this month in Cardiovascular Diabetology.

Two markers that measure different things

eGDR is a formula that combines waist circumference, whether you have high blood pressure, and HbA1c, and a lower score suggests more insulin resistance. It was checked against the gold-standard laboratory test of insulin sensitivity in people with type 1 and type 2 diabetes, but its accuracy in generally healthy adults without diabetes remains uncertain. For most people, then, it's probably best read as a compact summary of three familiar risk factors. Lp(a) is a cholesterol-carrying particle whose level is largely inherited, so it captures a different risk characteristic. Before this study, each had its own track record: lower eGDR was linked to more cardiovascular disease in a meta-analysis of cohort studies and in cohorts from China, England and the US, and high Lp(a) was independently linked to more events. An earlier UK Biobank analysis that paired Lp(a) with a different insulin-resistance measure also found people high on both had the highest risk.

UK Biobank is a long-running study of UK adults who gave blood samples and body measurements at enrollment and have been followed through health records since. Song Wen and colleagues analyzed 286,025 participants with no cardiovascular disease at the start. They defined low eGDR as 6.28 mg/kg/min or below and high Lp(a) as 125 nmol/L or above. The two markers were barely correlated, and each was linked to cardiovascular disease independently of the other, so a reassuring result on one tells you little about the other.

What having both looked like over a decade

Tests for interaction were not statistically significant on either the relative or the absolute scale, so the study did not establish that the markers amplify each other. Two other qualifications matter. These are group averages from a cohort of volunteers, so they aren't a personal forecast. And because the study is observational, the gap between groups is a measure of association, which is different from the amount of risk any treatment would remove.

Why the combined profile can't pick your treatment yet

The researchers also asked whether knowing both markers helps beyond a comprehensive conventional risk model. The model's ability to sort who would and wouldn't develop cardiovascular disease improved only in the third decimal place, and a decision analysis showed no clear gain in deciding whom to treat. One likely reason is that eGDR is built from waist size, hypertension status and HbA1c, which standard risk assessment already largely captures.

The evidence has other gaps. Independent prospective validation of these exact thresholds together is lacking, as is trial evidence that prevention guided by the combination improves on ordinary risk-guided care. The manuscript is an early accepted version, and adjustment for measured risk factors cannot rule out unmeasured confounding. Only the abstract and front matter were available for this review, leaving endpoint definitions and detailed confounding control unassessed.

What the evidence supports doing now

Lp(a) is the more actionable half. The 2026 ACC/AHA dyslipidemia guideline recommends measuring it at least once in adulthood and treats 125 nmol/L (or 50 mg/dL) and above as a risk-enhancing factor that should prompt more intensive LDL lowering and closer management of other risk factors. The 2025 European update takes a similar position above about 105 nmol/L. If you've never had your Lp(a) checked, it's worth doing. If it comes back high, how hard to push LDL down, and whether that means a prescription, is a decision to make with a clinician using your full risk picture, including the PREVENT risk estimate now recommended by the ACC/AHA guideline.

You don't need the eGDR formula to use what it measures. A low score means some combination of a larger waist, high blood pressure and higher HbA1c, and checking those directly helps identify what to work on. A low eGDR is not a diagnosis, and the study's cutoff is not a reason to start medication.

This study is a concrete reason to take both metabolic health and inherited lipid risk seriously, and elevated Lp(a) already has a defined place in prevention decisions. What would move the combined profile into routine care is an independent cohort confirming these thresholds and showing they improve real decisions, followed by a trial in which choosing treatment by the combination leads to fewer heart attacks and strokes with benefits that outweigh treatment harms.