A single combined measure of high-sensitivity C-reactive protein, low-density lipoprotein cholesterol, and lipoprotein(a) levels in initially healthy women predicted cardiovascular events over a 30-year period, according to a recent study.

In the study, published in The New England Journal of Medicine, investigators analyzed the data of 27,939 healthy U.S. women in the Women's Health Study, with a median follow-up of 27.4 years. During the study period, 3,662 first major adverse cardiovascular events (MACE) occurred.

The women involved in the study had a mean age of 54.7 years at baseline. At enrollment, 25.0% had hypertension, 12.0% were current smokers, 2.5% had diabetes, and 14.4% had a parental history of myocardial infarction before 65 years of age. The mean body mass index was 25.9.

The investigators measured high-sensitivity CRP, LDL cholesterol, and lipoprotein(a) levels at baseline. The primary endpoint of the study was a composite of myocardial infarction, coronary revascularization, stroke, or cardiovascular death.

Statistical analysis included calculation of adjusted hazard ratios (HR) across biomarker quintiles, along with 30-year cumulative incidence curves adjusted for age and competing risks. The investigators also performed joint-effect analyses of paired biomarkers and examined combined effects of all three biomarkers.

Among the key findings were:

• Increasing quintiles of baseline high-sensitivity C-reactive protein (CRP), low-density lipoprotein (LDL) cholesterol, and lipoprotein(a) all predicted 30-year cardiovascular risk.
• Covariable-adjusted HR for the primary endpoint (comparing highest to lowest quintiles): high-sensitivity CRP = 1.70 (95% confidence interval [CI] = 1.52–1.90), LDL cholesterol = 1.36 (95% CI = 1.23–1.52), lipoprotein(a) = 1.33 (95% CI = 1.21–1.47).
• Each biomarker showed independent contributions to overall risk.
• The greatest risk stratification was achieved using all three biomarkers in combination.

Biomarker ranges by quintile:

• High-sensitivity CRP (mg/L): Q1: < 0.65, Q2: 0.65 to < 1.47, Q3: 1.47 to < 2.75, Q4: 2.75 to < 5.18, Q5: ≥ 5.18
• LDL cholesterol (mg/dL): Q1: < 96.1, Q2: 96.1 to < 113.5, Q3: 113.5 to < 129.7, Q4: 129.7 to < 150.7, Q5: ≥ 150.7
• Lipoprotein(a) (mg/dL): Q1: < 3.6, Q2: 3.6 to < 7.6, Q3: 7.6 to < 15.5, Q4: 15.5 to < 44.1, Q5: ≥ 44.1.

After adjusting for covariables and other biomarkers, the HR for each increase in quintile were:

• High-sensitivity CRP: 1.14 (95% CI = 1.11–1.17)
• LDL cholesterol: 1.08 (95% CI = 1.05–1.10)
• Lipoprotein(a): 1.06 (95% CI = 1.03–1.08).

Covariable-adjusted and biomarker-adjusted HR for each standard deviation increase:

• High-sensitivity CRP: 1.23 (95% CI = 1.18–1.27)
• LDL cholesterol: 1.10 (95% CI = 1.06–1.14)
• Lipoprotein(a): 1.09 (95% CI = 1.05–1.13).

The effects of high-sensitivity CRP and LDL cholesterol on risk attenuated slightly over time. For high-sensitivity CRP, the HR per quintile increase was 1.17 (95% CI = 1.13–1.21) during 0 to 15 years of follow-up and 1.12 (95% CI = 1.09–1.16) during 15 to 30 years. For LDL cholesterol, the corresponding HR were 1.13 (95% CI = 1.09–1.17) and 1.06 (95% CI = 1.02–1.09), respectively.

Combined biomarker analysis revealed increasing risk with more biomarkers in the highest quintile. Covariable-adjusted HR were:

• One biomarker in quintile 5: 1.27 (95% CI = 1.19–1.37)
• Two biomarkers in quintile 5: 1.66 (95% CI = 1.51–1.83)
• Three biomarkers in quintile 5: 2.63 (95% CI = 2.16–3.19).

Similar trends were observed for individual endpoints of stroke and coronary heart disease.

Number of events by high-sensitivity CRP quintile:

• Q1: 455/5,659
• Q2: 586/5,575
• Q3: 695/5,539
• Q4: 841/5,582
• Q5: 1085/5,584

By year 30, 16,053 women (57.5%) reported having received at least one statin prescription. In sensitivity analyses censoring follow-up at first reported statin prescription, the predictive value of all three biomarkers remained strong. Covariable-adjusted HR for quintile 5 vs quintile 1 were:

• High-sensitivity CRP: 1.65 (95% CI = 1.43–1.90)
• LDL cholesterol: 1.62 (95% CI = 1.41–1.86)
• Lipoprotein(a): 1.42 (95% CI = 1.25–1.62).

Subdistribution HR from Fine-Gray models were similar to the main analysis.

The study limitations included the predominantly White female cohort and lack of repeated biomarker measurements.

Disclosures can be found in the study.