Meta-analysis of transcatheter aortic valve implantation for bicuspid versus tricuspid aortic valves

Hisato Takagi; Yosuke Hari; Norikazu Kawai; Toshiki Kuno; Tomo Ando; ALICE (All-Literature Investigation of Cardiovascular Evidence) Group

doi:10.1016/j.jjcc.2019.03.018

Meta-analysis of transcatheter aortic valve implantation for bicuspid versus tricuspid aortic valves

J Cardiol. 2019 Jul;74(1):40-48. doi: 10.1016/j.jjcc.2019.03.018. Epub 2019 Apr 21.

Authors

Hisato Takagi¹, Yosuke Hari², Norikazu Kawai³, Toshiki Kuno⁴, Tomo Ando⁵; ALICE (All-Literature Investigation of Cardiovascular Evidence) Group

Affiliations

¹ Department of Cardiovascular Surgery, Shizuoka Medical Center, Shizuoka, Japan; Department of Cardiovascular Surgery, Kitasato University School of Medicine, Sagamihara, Japan. Electronic address: kfgth973@ybb.ne.jp.
² Department of Cardiovascular Surgery, Shizuoka Medical Center, Shizuoka, Japan; Department of Cardiovascular Surgery, Kitasato University School of Medicine, Sagamihara, Japan.
³ Department of Cardiovascular Surgery, Shizuoka Medical Center, Shizuoka, Japan.
⁴ Department of Medicine, Mount Sinai Beth Israel Medical Center, New York, NY, USA.
⁵ Department of Cardiology, Detroit Medical Center, Detroit, MI, USA.

PMID: 31015000
DOI: 10.1016/j.jjcc.2019.03.018

Abstract

Background: We performed meta-analysis and meta-regression of transcatheter aortic valve implantation (TAVI) for the bicuspid aortic valve (B-AV) versus the tricuspid aortic valve (T-AV).

Methods: MEDLINE and EMBASE were searched through June 2018 using PubMed and OVID. We included comparative studies of TAVI patients with B-AV versus T-AV reporting at least one of postprocedural transcatheter valve regurgitation (TVR)/pacemaker implantation (PMI) incidence and early (30-day or in-hospital)/late (including early) mortality. For each study, crude (unadjusted) data regarding TVR/PMI incidence and early/late mortality in both the B-AV and T-AV groups were used to generate risk ratios (RRs). Study-specific estimates were combined in the random-effects model. Using meta-regression, we assessed potential confounders identified in preliminary meta-analysis.

Results: We identified 12 eligible studies including a total of 1045 B-AV and 4069 T-AV patients. Pooled analysis demonstrated an association of B-AV with a statistically significant increase in TVR incidence (RR, 1.42; p=0.006) but no statistically significant difference in PMI incidence (p=0.54) and 30-day (p=0.11)/midterm (1-year to 2-year) mortality (p=0.99) between patients with B-AV and those with T-AV. All meta-regression coefficients of 6 identified potential confounders (age, mean aortic valve gradient, aortic valve area, left ventricular ejection fraction, aortic calcification, and B-AV types) for the outcomes (TVR/PMI incidence and early/late mortality) were statistically non-significant.

Conclusions: Postprocedural PMI incidence and 30-day/midterm (1-year to 2-year) mortality after TAVI may be similar between patients with B-AV and those with T-AV despite the significant association of B-AV with increased postprocedural TVR incidence.

Keywords: Bicuspid aortic valve; Meta-analysis; Meta-regression; Transcatheter aortic valve implantation; Tricuspid aortic valve.

Publication types

Meta-Analysis

MeSH terms

Aged
Aged, 80 and over
Aortic Valve / abnormalities*
Aortic Valve / pathology
Aortic Valve / surgery
Bicuspid Aortic Valve Disease
Heart Valve Diseases / surgery*
Humans
Incidence
Male
Odds Ratio
Postoperative Complications / etiology
Postoperative Complications / mortality*
Risk Factors
Stroke Volume
Transcatheter Aortic Valve Replacement / mortality*
Treatment Outcome
Tricuspid Valve / surgery*
Ventricular Function, Left