Long-term risk of a major cardiovascular event by apoB, apoA-1, and the apoB/apoA-1 ratio-Experience from the Swedish AMORIS cohort: A cohort study

Abstract

Background: Elevated apolipoprotein B (apoB) and elevated apoB/apoA-1 ratio increase the risk of myocardial infarction (MI) and stroke, whereas high apoA-1 is protective. We study how these apolipoproteins are associated with major adverse cardiovascular events (MACEs), whether apoA-1 contributes to this association, and whether abnormal values occur decades before such events develop.

Methods and findings: In the Swedish AMORIS (Apolipoprotein-related MOrtality RISk) cohort study, 137,100 men and women aged 25-84 years were followed an average 17.8 years. ApoB, apoA-1, and the apoB/apoA-1 ratio were analysed in relation to MACEs (non-fatal MI, stroke, and cardiovascular [CV] mortality), yielding 22,473 events. Hazard ratios (HRs) were estimated using Cox regression. Kaplan-Meier estimates were used to investigate the relationship of MACEs with increasing quintiles of the apoB/apoA-1 ratio in all age groups for both sexes. In nested case-control analyses, cases were randomly matched to age- and sex-matched controls, yielding population trajectories for apolipoproteins. Increased level of apoB and increased apoB/apoA-1 ratio were associated with risk of MACE and all clinical sub-components in both men and women across all ages (10th versus first decile in both sexes combined: HR 1.7 for MACE and 2.7 for non-fatal MI). Decreased values of apoA-1 potentiated the impact of apoB at all levels of apoB (on average across apoB range: 40% increase in HR for MACE and 72% increase in HR for non-fatal MI), indicating that the apoB/apoA-1 ratio covers a broader range of persons with dyslipidaemia at risk than apoB alone. In both men and women, MACEs occurred earlier on average for each increasing quintile of the apoB/apoA-1 ratio. Individuals with the highest levels of apoB/apoA-1 ratio experienced CV events on average several years earlier than those with lower ratios. Higher apoB/apoA-1 ratio in cases of MACE versus controls was seen already about 20 years before the event. A limitation of this study was that adjustment for tobacco smoking and hypertension was only possible in a small validation study.

Conclusions: An imbalance between apoB and apoA-1 resulting in an increased apoB/apoA-1 ratio is strongly associated with the outcome MACE and its sub-components, in both men and women of all ages. An increased apoB/apoA-1 ratio already 2 decades before events calls for early recognition and primary prevention. Simple evidence-based cut values should be considered in future cardiovascular guidelines.

Fig 1. Kaplan–Meier estimate for quintiles (Q1–Q5, highest) of the apoB/apoA-1 ratio (apo-ratio) for major adverse cardiovascular events in men and women.

Fig 2. Hazard ratios of cardiovascular outcomes, with 95% confidence intervals, for deciles of the apoB/apoA-1 ratio.

Men and women combined. Adjusted for total cholesterol, triglycerides, glucose, sex, and socioeconomic status. Outcomes: (a) major adverse cardiovascular event (MACE); (b) myocardial infarction; (c) ischaemic stroke; (d) cardiovascular mortality; (e) MACE, coronary artery bypass grafting (CABG), or percutaneous coronary intervention (PCI); (f) CABG or PCI.

Fig 3. Hazard ratios of cardiovascular outcomes in AMORIS and WOLF/60YO, with 95% confidence intervals, for quintiles of the apoB/apoA-1 ratio.

Men and women combined. Adjusted for total cholesterol, triglycerides, glucose, sex, and SES and, as noted in the figure, for hypertension, diabetes, and smoking. AMORIS: n = 137,100 (health check-up, n = 39,007); WOLF/60YO: n = 13,636. To enhance readability, the blue confidence intervals have been slightly shifted to the left, so as not to overlap with the red intervals. Outcomes: (a) MACE; (b) myocardial infarction; (c) ischaemic stroke; (d) cardiovascular mortality; (e) MACE, CABG, or PCI; (f) CABG or PCI. 60YO, Cohort of 60-year-olds; CABG, coronary artery bypass grafting; MACE, major adverse cardiovascular event; PCI, percutaneous coronary intervention; WOLF, Work, Lipids and Fibrinogen.

Fig 4. Hazard ratios of major adverse cardiovascular event (MACE) and myocardial infarction by the combination of deciles for apoA-1 (g/L) and apoB (g/L).

Men and women combined. Adjusted for total cholesterol, triglycerides, glucose, sex, and socioeconomic status. MACE, top; myocardial infarction, bottom.

Fig 5. ROC-AUC for the apoB/apoA-1 ratio and apoB for different outcomes for men and women aged 25–69 years.

CABG, coronary artery bypass grafting; CV-Mort, cardiovascular mortality; MACE, major adverse cardiovascular event; MI, myocardial infarction; PCI, percutaneous coronary intervention; ROC, receiver operating characteristic curve.

Fig 6. Trajectories of apoB/apoA-1 ratio, apoB, and apoA-1 over a 25-year period prior to a major adverse cardiovascular event (MACE) or myocardial infarction, for cases and controls, in men and women combined.

MACE (left); myocardial infarction (right). ApoB/apoA-1 ratio (top), apoB (middle), and apoA-1 (g/L) (bottom). Note the different scales (y-axis) for the levels of the apolipoproteins (g/L).