Comparison of imaging vs. plasma biomarkers of vascular inflammation for cardiovascular prognostication: a meta-analysis

Abstract Funding Acknowledgements Type of funding sources: None. Background Measurement of inflammatory biomarkers is advocated for cardiovascular risk stratification, but to date there is no systematic evaluation of the incremental value of such approaches and their comparative performance for cardiovascular risk prediction. Methods We systematically explored published literature in MEDLINE before Apr 14, 2020 for prospective cohort studies assessing the prognostic value of common biomarkers of vascular inflammation in stable patients with or without cardiovascular disease. The primary outcome was the difference in the reported c-index (Δ[c-index]) of the best clinical model with the use of inflammatory biomarkers for the prediction of the composite of major adverse cardiovascular events (MACEs) and mortality. The secondary outcome was the Δ[c-index] for MACEs only. We calculated I² to test heterogeneity. We used random-effects modelling for the meta-analyses to assess the primary and secondary outcome. Results We identified 94,821 studies in MEDLINE, of which 90,882 (96%) studies were excluded after screening the titles and abstracts, and 1,507 (93%) of the 1,625 remaining studies were excluded after assessment of the full texts. We included 93 (6%) studies in our quantitative evaluation, which comprised 351,628 individuals. The median c-index and Δ[c-index] for the composite endpoint per biomarker type are shown in panels A & B. In meta-analysis, plasma or imaging biomarkers of vascular inflammation offered incremental prognostic value for the primary outcome (pooled estimate for Δ[c-index]% 2.9, 95%CI 2.1-3.6, I2 = 72%) and for MACEs only (pooled estimate for Δ[c-index]% 2.9, 95%CI 2.1-3.8, I2 = 74%, panel C). The prognostic value of imaging biomarkers significantly outperformed that of plasma biomarkers for the primary outcome (panel D, Δ[c-index]% 11.3, 95%CI 8.3-14.3 vs. 1.4, 95%CI 0.9-1.8 respectively, p = 1.7x10-10). In other subgroup analyses (panel E), biomarkers of vascular inflammation performed better in studies with a shorter duration of follow-up (i.e. <5 years), in primary CHD prevention setting and lower cardiovascular risk populations i.e. (studies with <5% cumulative event incidence). Conclusions The use of plasma, but mainly imaging, biomarkers to detect vascular inflammation provides incremental prognostic value over clinical models for cardiovascular events. Implementation of such approaches in clinical practice could improve cardiovascular risk prognostication and reduce cardiovascular disease burden. Abstract Figure.