Background: Aortic stenosis (AS) can cause cardiogenic shock (CS). Transcatheter aortic valve replacement (TAVR) is the standard of care for elderly and severely ill patients with AS and is well established in many hospitals. However, it’s use in patients with CS due to AS needs further investigation. Despite increasing safety of the procedure, complication rates in these patients are high. The aim of this study was to explore clinical characteristics and long-term survival of patients treated with emergency TAVR. We hypothesized that patients who survive the procedure and initial intensive care therapy have a prognosis comparable to patients with stable AS undergoing elective TAVR.

Methods: All patients undergoing TAVR between 2013 and 2019 at our centre were screened. Patients with CS or severe decompensated due to AS treated with emergency TAVR or requiring intensive care medicine before the procedure were included in the analysis. Patients were dived into two groups according to the presence or absence of CS, defined as systolic blood pressure of < 90 mmHg for 30 minutes or need for vasopressors and an elevated serum lactate of ≥ 2 mmol/L.

Results: In total, 189 patients were analysed, of which 53 patients were found to be in CS (shock group) and 136 patients did not fulfil shock criteria. The remaining patients undergoing elective TAVR served as a comparison cohort.  Patients in the shock group were more often male (75.5% vs. 56.6%, p=0.02), had a higher Society of Thoracic Surgeons score (15.6 [IQR, 7.7-30.5] vs. 5.4 [3.8-8.3], p<0.01) and a lower leftventricular ejection fraction (38.0% vs. 48.0%, p<0.01). The rate of coronary artery disease was similar (78.0% vs. 75.2%, p=0.69). On admission, more shock patients had elevated lactate (84.3% vs. 41.9%, p<0.01), and decreased blood pressure (57.7% vs. 7.4%, p<0.01). Concerning the cause of acute decompensation, a specific condition which resulted in a clinical deterioration triggering hospitalisation could be identified in most patients. These included, among others, volume overload, acute coronary syndromes, arrhythmias, infections, and bleeding. Before TAVR, patients were treated on an ICU for 3 [2-5] days (shock group) and 2 [1-5] days (urgent group), p=0.01. In the shock group, mechanical ventilation was more frequent (52.8 vs. 14.7%, p<0.01), and more patients received valvuloplasty before TAVR (18.9 vs. 3.7%, p<0.01) (Table). The Valve Academic Research Consortium 3 (VARC-3) composite endpoint of technical failure occurred more often in emergency patients (shock, 13.2% vs. urgent, 9.6%) than in the 2,741 patients after elective TAVR (4.8%, p<0.01). Estimated 2-year mortality was higher in the shock compared to urgent group (hazard ratio, 2.1 [1.3-3.3], p<0.01). In a landmark analysis from day 90, 2-year mortality in these groups was comparable to elective TAVR patients (p=0.41, Figure).

Conclusion: Procedural complications are more common in emergency TAVR than in elective ones. Mortality in patients with cardiogenic shock is higher than in other emergency TAVR patients. If the first 90 days after emergency TAVR were survived, TAVR patients had a prognosis similar to elective TAVR patients.