Does diabetes mellitus impact prognosis after transcatheter aortic valve implantation? Insights from a meta-analysis

Abstract

Background: Diabetes mellitus (DM) is well known to increase mortality in several cardiovascular diseases. However, the prognostic impact of DM following transcatheter aortic valve implantation (TAVI) remains controversial. We sought to assess the impact of DM on perioperative (in-hospital or 30-day) complications and mid-term (≥1 year) all-cause mortality after TAVI through meta-analysis.

Methods: A comprehensive literature search of PUBMED and EMBASE was conducted through January 1st 2002 to May 15th 2016. Articles that reported adjusted hazards ratio (HRs) or unadjusted HR for mid-term all-cause mortality with 95% confidence intervals (CIs) of DM or insulin dependent DM (IDDM) on mid-term all-cause mortality post TAVI were included in the analysis. A meta-analysis was performed with combination of both adjusted HR and un-adjusted HR. Sensitivity analysis was performed with only the adjusted HR. Random-effects model was used to calculate the pooled effect size.

Results: A total of 22 observational cohort studies were identified with 28,440 (8998 DM and 19,442 non-DM) patients. The risk of perioperative complications (myocardial infarction, bleeding, major vascular complications, stroke, and new-onset atrial fibrillation) was similar between DM and non-DM cohorts. A meta-analysis of all-cause mortality of DM (19 studies after excluding 3 studies that only reported HR of IDDM on mid-term all-cause mortality, 8808 DM and 17,829 non-DM patients) resulted in significantly worse outcome (HR 1.21, 95%CI 1.10-1.34, p=0.0002, I²=53%) in DM patients compared to non-DM patients post-TAVI. Sensitivity analysis showed consistent results. Subgroup analysis (4 studies with 267 IDDM versus 2161 non-IDDM) demonstrated that IDDM was associated with higher all-cause mortality (HR 2.05, 95%CI 1.54-2.73, p<0.00001, I²=0%) following TAVI.

Conclusions: DM was associated with similar perioperative complications but was associated with increased mid-term all-cause mortality after TAVI. Further study of the causes of increased mortality during the follow-up may lead to improved outcome.

Keywords: Diabetes mellitus; Meta-analysis; Mortality; Transcatheter aortic valve implantation.