Residual heart attack risk persists after LDL drops
Lowering LDL cholesterol remains one of the most effective tools in cardiovascular prevention, yet a lower number does not erase every threat inside the arterial wall. New work from Chalmers University of Technology points to a reason some patients still go on to develop coronary heart disease even after treatment brings bad cholesterol into the recommended range: other apoB-containing lipoproteins may drive inflammation and vascular injury through pathways that go well beyond simple cholesterol deposition.
The study, published in the European Heart Journal, focused on three major fat-carrying particles in blood: LDL, triglyceride-rich lipoproteins, or TRL, and lipoprotein(a), or Lp(a). All three can contribute to atherosclerosis, but they do not appear to behave identically in the body. LDL mainly tracks with cholesterol accumulation in the artery wall. TRL and Lp(a) appear to do something more complex, with protein signatures that point toward immune-cell recruitment, inflammation, and structural changes in blood vessels.
That distinction matters at the bedside. Patients can reach target LDL levels and still carry a meaningful burden of atherogenic particles capable of irritating the vascular lining.
Proteomic signals point to inflammatory damage beyond cholesterol load
To move beyond cholesterol counting, investigators used proteomics analysis to examine nearly 3,000 proteins in blood samples, then compared those patterns with genetic information and long-term health records. Proteomics offers a functional snapshot of biology at a given moment, making it well suited to detect the molecular effects of lipoproteins that may not be obvious from standard lipid testing alone. Instead of asking only how much cholesterol each particle carries, the analysis asked what biological processes those particles seem to trigger.
The overlap between TRL and Lp(a) stood out. Their associated protein patterns were strikingly similar, and those shared signatures tracked with later heart disease. The proteins linked to those particles were tied to attracting immune cells and to changes in the structure of vessel walls, a combination that fits with ongoing arterial injury rather than passive fat accumulation alone. In practical terms, the findings suggest that these particles may help sustain a smoldering inflammatory state inside the arteries even when LDL has already been brought down.
LDL produced a different signature. Its associated proteins were more closely linked to cholesterol buildup in artery walls, reinforcing the long-established role of LDL in plaque formation while separating it biologically from the effects seen with TRL and Lp(a). The result is a sharper picture of atherosclerosis: not one uniform pathway, but multiple lipoprotein-driven mechanisms that converge on the same clinical endpoint.
TRL and Lp(a) may share a vascular injury pathway
TRL and Lp(a) deserve close attention because they may carry a disproportionate share of residual risk. TRL includes triglyceride-rich particles such as liver-derived VLDL and the remnants left behind as triglycerides are cleared. Those remnants can still deposit cholesterol in artery walls. Lp(a), meanwhile, resembles LDL but carries an additional apolipoprotein(a) component and is already recognized as an established risk factor for coronary heart disease.
The biology matters because atherosclerosis is not simply a storage problem. Once lipoproteins enter the arterial wall, they can ignite local immune responses, alter endothelial behavior, and thicken or destabilize plaques over time. If TRL and Lp(a) are amplifying that process through shared inflammatory and vascular pathways, then standard LDL lowering may leave part of the disease biology untouched. That would help explain why a subset of patients remains vulnerable despite apparently successful lipid management.
A second direct consequence follows from the same logic: risk prediction based only on LDL can miss important biology.
The group’s earlier work had already suggested that counting harmful lipoprotein particles predicted future heart disease better than cholesterol concentration alone. This newer proteomic analysis extends that idea by showing why particle number and particle type may matter. Two people can arrive at the same LDL value and still differ in the composition and inflammatory behavior of the particles circulating in their blood.
Prevention may need to move beyond one lipid target
Clemens Wittenbecher, who led the study, framed the findings around a familiar clinical truth: lowering LDL remains a cornerstone of prevention, but it does not fully explain the risk that persists after treatment. That point is difficult to overstate in cardiovascular medicine. LDL reduction works, but atherosclerosis is broader than LDL alone, and the blood vessel wall responds to multiple overlapping metabolic signals.
The current study does not establish a new treatment or a blood test ready for routine use. It also does not show that measuring these protein patterns will improve outcomes in everyday practice. What it does provide is a plausible biological map for future intervention. If TRL and Lp(a) are driving vascular injury through inflammatory pathways, then therapies that specifically lower these particles or blunt their downstream effects could close part of the gap left by LDL-focused care.
For now, the clinical message is conservative but important. LDL control remains essential. Patients with continued risk despite optimal LDL reduction may be carrying additional burden from TRL, Lp(a), or both, and those particles may be acting through mechanisms that standard cholesterol testing cannot fully capture. The next step is to determine whether measuring or targeting these processes changes hard outcomes such as heart attacks and vascular death, not just laboratory values.
Until that evidence arrives, the study leaves cardiovascular prevention in a more biologically precise place, but not a finished one. The mechanism is clearer than before, yet the therapeutic playbook is still incomplete.