Prognostic impact of baseline C‐reactive protein levels on mortality after transcatheter aortic valve implantation
Hisato Takagi, Toshiki Kuno, Yosuke Hari, Kouki Nakashima, Yujiro Yokoyama, Hiroki Ueyama, Tomo Ando, ALICE (All‐Literature Investigation of Cardiovascular Evidence) Group
Abstract
OBJECTIVES: To determine whether baseline C-reactive protein (CRP) levels can predict mortality after transcatheter aortic valve implantation (TAVI), we performed a meta-analysis of currently available studies. METHODS: All studies investigating the prognostic impact of baseline (preprocedural) CRP levels on all-cause mortality after TAVI were identified by means of searching PubMed and Google Scholar through May 2019. For each study, (preferentially, adjusted rather than unadjusted) odds/hazard ratios (ORs/HRs) with corresponding 95% confidence intervals of mortality per standard-deviation (SD) (or unit) increase in CRP levels or those for high vs low CRP levels. RESULTS: Our search identified 14 eligible studies including a total of 3449 patients undergoing TAVI and reporting early (in-hospital to 3-month) and midterm (1-year to 3-year) all-cause mortality after TAVI. Pooled analyses demonstrated associations of high-baseline CRP levels with a marginal, but statistically nonsignificant increase in early mortality (pooled OR/HR per SD increase in CRP levels, 2.72; P = .09 and pooled OR/HR for high vs low CRP levels, 3.32; P = .07) and a statistically significant increase in midterm mortality after TAVI (pooled OR/HR per SD increase in CRP levels, 1.45; P < .0001 and pooled OR/HR for high vs low CRP levels, 1.78; P < .00001). Excluding HRs for high-sensitivity CRP, combining ORs/HRs of 1-year mortality, pooling HRs of ≥2-year mortality, and combining adjusted HRs did not alter the primary results. CONCLUSION: High-baseline CRP levels may predict increased midterm, but not early, mortality after TAVI.