External validation of isolated PE-related proteins in a prognostic, population-based setting. This figure shows the hazard ratios and 95% CI per SD increase in protein concentration for all isolated PE-related proteins that could be externally validated (18/70), in the prospective, population-based cohort study the Gutenberg Health Study. Individual Cox proportional hazards regression models were used to identify the relationship between each protein and the formation of a first incident isolated PE in a population-representative sample free of history of VTE and active cancer, adjusted for age, sex, BMI, cardiovascular disease, COPD, cardiovascular risk factors and intake of antithrombotic medication, NSAIDs, corticosteroids, and oral contraceptives. P values with asterisks signify those that survived Bonferroni multiple testing correction (ie, those P values below a 2-sided significance threshold of 0.05/18 = 0.0028). NSAID, nonsteroidal anti-inflammatory drug.