Readmission following urgent transcatheter aortic valve implantation versus urgent balloon aortic valvuloplasty in patients with decompensated heart failure or cardiogenic shock

2021 
BACKGROUND Urgent transcatheter aortic valve implantation (TAVI) is a feasible option for aortic stenosis (AS) patients with decompensated heart failure (HF) and cardiogenic shock (CS) as compared to the more traditional urgent balloon aortic valvuloplasty (BAV). OBJECTIVES We conducted a retrospective analysis to compare risk and cause of readmission in these two high-risk groups. METHODS Nationwide Readmission Database (NRD) 2011-2014 was retrospectively analyzed to identify patients with AS having either urgent TAVI or urgent BAV using appropriate ICD-9 codes. Propensity scores were used to match patients with urgent TAVI as compared to patients with urgent BAV. Statistical analysis was performed using the Stata 15.1 software. RESULTS We identified a weighted sample of 6,670 patients with urgent BAV and 6,964 patients with urgent TAVI. The all-cause 30- and 90-day readmission was lower in the urgent TAVI group compared to urgent BAV (15.4 vs. 22.5%, (aHR): 0.92 [0.90-0.95] p < .001). 30-day readmission due to CV cause and HF was also lower in the urgent TAVI group (aHR, 0.93: p < .001 and aHR, 0.98: p = .040, respectively). The 30-day gastrointestinal (GI) bleed readmission rate was three times higher in urgent TAVI group (aHR, 3.00:95% CI (1.23-7.33), p = .016), but was not statistically significant at 90-days. Cardiac causes of readmission were the predominant cause of readmission in both groups, but more pronounced in urgent BAV group (60.3 vs. 40.5%, p < .001). CONCLUSION Urgent TAVI appears beneficial in patients with AS and decompensated HF or CS driven by roughly 10 and 25% reductions in overall readmissions at 30 and 90 days, and marked reductions in reintervention, although offset partially by higher risk of readmission due to GI bleeding at 30 days.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    9
    References
    1
    Citations
    NaN
    KQI
    []