C-reactive protein levels and coronary in-stent restenosis: A systematic review and meta-analysis

Abstract

Abstract Background C-reactive protein (CRP) is an inflammatory biomarker and has been implicated in the pathogenesis of atherosclerotic lesion formation, plaque rupture, and coronary thrombosis. Its relationship to adverse angiographic outcomes after coronary stenting however remains uncertain. Purpose We aimed to test the association of baseline CRP levels with the incidence of coronary in-stent restenosis (ISR) using a systematic review and meta-analysis. Methods Relevant published studies were identified following a US National Institutes of Health (PubMed), EMBASE, MEDLINE, Scopus and Web of Knowledge databases search, until 31st October 2022. Selected studies had to be/have (i) original research with an angiographically determined ISR and an angiographically or clinically determined no-ISR group, and (ii) CRP levels measured before index stenting for both groups. Mean difference in CRP levels with its 95%CI between ISR and no-ISR was ascertained for each included study. Pooled standard mean difference (SMD) and its 95%CI was derived after pooling study level results using random effects employing a Z test. Begg’s funnel plots and Egger’s test were used for publication bias assessment. Results Amongst total of 1,009 unique results, 497 full-text articles were assessed for eligibility, 461 were excluded for a variety of reasons, which left 36 articles with 37 different studies (Nst= 37), with a total sample of 10,577 (2,122 ISR cases / 8,455 controls), to be included for quantitative synthesis. Complete study flow is depicted in Figure 1. ISR group had higher pooled CRP levels (Nst= 37, SMD= 0.50, 95% CI= 0.33, 0.67 mg/L, p< 0.00001) as compared to no-ISR group. Sensitivity analysis found no instance of significant alteration from the original results. No evidence of publication bias was detected either visually by the Begg’s funnel plots or the Egger’s test (p= 0.07). Analysis was repeated after subgroup stratification of the included studies according to study design, stent type, assay type, ancestry, presentation diagnosis, and data type. Important findings identified by the subgroup analysis were the significant differences in CRP levels between ISR and no-ISR, amongst Middle Easterners (Nst= 10, SMD 0.63, 95%CI 0.33, 0.93 mg/L, p< 0.0001) and Asians (Nst= 18, SMD 0.71, 95%CI 0.46, 0.96 mg/L, p= 0.00001), however not amongst Europeans (Nst= 9, SMD -0.11, 95%CI -0.43, 0.21 mg/L, p= 0.50). Results obtained for other subgroups however remained consistent and showed significant positive associations. (Figure 2.) Conclusions The present systematic review and meta-analysis indicates a positive association of higher baseline CRP values with the incidence of angiographically confirmed ISR. In subgroup analysis there was no evidence of association amongst Europeans, in contrast to significant associations seen amongst rest of the ancestries. These results warrant further validation in dedicated large cohorts, ideally in a prospective setting.Figure 1.Figure 2.