As you like association studies @Joseph_Lavelle : Novel Classification of Cardiovascular Disease Subtypes Reveals Associations Between Mortality and Polyunsaturated Fatty Acids: Insights from the United Kingdom Biobank Study 2024
They clustered people into 3 CVD subtypes based on some biomarkers:
Principal component analysis and k-means clustering were used to determine the CVD subtype. Variables included age, body mass index, waist–hip ratio, diastolic blood pressure, systolic blood pressure, total cholesterol, total triglycerides, high-density lipoprotein-cholesterol, apolipoprotein B:apolipoprotein A1, glycated hemoglobin, creatinine, albumin, C-reactive protein, white blood cell count, platelet count, and hemoglobin concentration.
Three distinct CVD subtypes were identified, with cluster 3 characterized by older age, male gender, and low high-density lipoprotein-cholesterol, having the highest risk of mortality. Clusters 2 and 3 had the highest DHA and ω-6/ω-3 ratios, respectively, compared with Cluster 1.
Cluster 1 consisted of the youngest individuals with elevated levels of ApoB:ApoA1, TC, TG, SBP, and DBP. Cluster 2 predominantly comprised females (77.9%) with the lowest BMI, WHR, TG, creatinine, C-reactive protein, WBC, and hemoglobin concentration, while having the highest levels of HDL-cholesterol. Cluster 3 primarily consisted of males (72.7%) who were older, with a higher BMI, WHR, and HbA1c, and a lower TC, HDL-cholesterol, and PLT.
They adjusted for:
- sociodemographic factors, including gender and household income (less than £18,000, £18,000 to £30,999, £31,000 to £51,999, £52,000 to £100,000 and greater than £100,000);
- socioeconomic status, including Townsend deprivation index;
- lifestyle habits, including smoking status (never, former, and current), alcohol status (never, former, and current), and physical activity (low, moderate, and high);
- comorbidities, including hypertension and diabetes; 5) drug use, including cholesterol-lowering medication use, antihypertensive drugs use, insulin treatment, and aspirin use.
Results:
In multivariate models, we found significant nonlinear relationships between total PUFAs, ω-3, and DHA and risk of all-cause, CVD, and IHD mortality (all P for nonlinearity < 0.05). There was an inverse L-shaped exposure–response relationship between these PUFAs and risk of all-cause mortality, whereas there was a U-shaped exposure–response relationship for risk of CVD and IHD mortality. In the absence of these PUFAs, risk of the outcome increases. Within a specific range, higher ω-3 and DHA concentrations are associated with a reduced risk of outcomes [hazard ratio (HR) < 1.0]. However, once PUFAs reach a specific threshold level, risk stabilizes and no longer continues to decrease or even increase. Cluster 1 displayed the lowest HR under similar exposure levels in all-cause mortality (Figure 10).
We found significant linear relationships between ω-6 and LA and risk of the studied outcomes (all P for linearity < 0.05). High level of these PUFAs were associated with low risk of all-cause and CVD mortality but were associated with an increased risk of IHD mortality.
Our findings suggest that the mortality rates of ω-3 and DHA with CVD and IHD begin to trend upward at high intakes (fourth quartile array). The possible reason for this is that excess ω-3 and DHA may be predisposing factors for atrial fibrillation
So, even this association study confirms that higher might not be better, especially for DHA. The potential inconsistency with previous studies is, I guess, that this one adjusted for many factors, especially income and education. Rich people eat more fish, and if you don’t account for that, the omega-3 index is just a wealth index. Physical exercise might also help to absorb omega 3, so it’s good that this study adjusted for “physical activity (low, moderate, and high)”.
You also asked “Do these results apply to healthy people?”: here they “constructed a subtype model of cardiovascular patients” “according to patient baseline characteristics”. They identified 3 clusters that I would call “young unhealthy” (1), “healthy” (2, mostly females), and “old unhealthy” (3). The results were similar across these clusters.
