Summary: In a secondary lipidomic analysis of two randomized, controlled, crossover feeding trials (DELTA 1, 96 healthy adults; DELTA 2, 79 metabolically challenged adults), reducing dietary saturated fat reshaped the lipoprotein(a) [Lp(a)] lipidome, with 87 of 440 lipid species differing significantly between diets (triacylglycerols changing most), yet plasma Lp(a) concentrations rose rather than fell. This is a mechanistic, associational finding with no cardiovascular outcomes.
PICO Summary
| Element | Detail |
|---|---|
| Population | Two multicenter, randomized, controlled, crossover feeding trials: DELTA 1 (n=96 healthy adults) and DELTA 2 (n=79 metabolically challenged adults). United States. Secondary lipidomic analysis of stored samples. |
| Intervention | Controlled diets with reduced saturated fatty acids (SFA), with SFA replaced by carbohydrate or unsaturated fat, fed within the crossover design. |
| Comparison | Higher-SFA average American control diet within the same participants (crossover), so each person served as their own comparator. |
| Outcome | Of 440 annotated Lp(a) lipid species, 87 differed significantly between diets (p<0.05 adjusted for multiplicity), triacylglycerols showing the most pronounced changes; TAG species with higher carbon and double-bond counts increased most with SFA reduction. Total Lp(a)-OxPL and four major OxPL subspecies showed no between-diet difference in DELTA 1; ALDOPC decreased significantly in DELTA 2 when SFA was replaced with carbohydrate (p=0.014). Plasma Lp(a) concentrations increased with SFA reduction (as reported in the parent trials). No 95% CI, ARR, NNT, or clinical outcomes reported. |
Expert Commentary
This is a hypothesis-generating lipidomic study, not evidence that diet changes cardiovascular risk through Lp(a). The verdict is that reducing saturated fat measurably remodels the Lp(a) lipidome, dominated by shifts in triacylglycerol species, while plasma Lp(a) is nudged upward rather than downward. That dissociation is biologically interesting but clinically ambiguous, because a higher Lp(a) concentration is generally unfavourable, whereas the compositional shifts have no established outcome correlate. The headline numbers are honest and modest: 87 of 440 species changed after multiplicity adjustment, and only one oxidised phospholipid subspecies moved, in only one cohort and only in the carbohydrate-replacement arm. The principal limitation is that this is a secondary analysis of two feeding trials, so it is associational and mechanistic; it was not designed or powered to link lipidome changes to events, and the small, specialised cohorts limit generalisability. Can I use this with my patients? Not yet. There is nothing here to change dietary counselling for a patient with elevated Lp(a), and it would be wrong to imply that cutting saturated fat lowers Lp(a) risk, since concentrations rose. What this work does is map an underexplored compartment and invite the obvious next step. Future studies should test whether these compositional shifts track with hard cardiovascular outcomes before any clinical inference is drawn. No commercial sponsorship distorts the message, and the effect sizes are plausible rather than implausibly large.
References
Myagmarsuren M, Law HG, Zhang W, Anuurad T, Bang H, Bishop LM, et al. Lipoprotein(a) lipidome: responses to reduced dietary saturated fat intake in two randomized controlled feeding trials. Nutrients. 2025;17(19):3113. doi:10.3390/nu17193113
