1Universitäres Herz- und Gefäßzentrum Hamburg Klinik für Kardiologie Hamburg, Deutschland; 2Wissenschaftliches Institut der AOK Berlin, Deutschland; 3Wissenschaftliches Institut der AOK 10178, Deutschland; 4HELIOS Klinikum Berlin-Buch Klinik und Poliklinik für Kardiologie und Nephrologie Berlin, Deutschland; 5Charité - Universitätsmedizin Berlin Klinik für kardiovaskuläre Chirurgie Berlin, Deutschland; 6Herz- und Diabeteszentrum NRW Klinik für Thorax- und Kardiovaskularchirurgie Bad Oeynhausen, Deutschland; 7Charité - Universitätsmedizin Berlin Notfall- und Akutmedizin Berlin, Deutschland; 8GKV-Spitzenverband (National Association of Statutory Health Insurance Funds) Berlin, Deutschland
BACKGROUND
Transcatheter aortic valve implantation (TAVI) has become the treatment modality of choice for the majority of patients with severe aortic stenosis. However, only limited data on the need for aortic valve (AV) reintervention during a longer follow up period is available.
AIM
We sought to determine the prevalence and mode of AV reintervention as well as its prognostic impact among patients who underwent TAVI in an all-comers patient population based on German administrative claims data.
METHODS
Data from patients undergoing TAVI between 2008 and 2018 from Germany's largest statutory health-care insurance provider were analyzed. AV reinterventions, namely redo surgical aortic valve replacement (re-SAVR), redo TAVI (re-TAVI), and redo balloon aortic valvuloplasty (re-BAV), were identified using OPS-codes, defined as any AV procedure beyond 30 days post-initial intervention.
A 3:1 propensity score matching for age and common comorbidities was performed between patients receiving re-TAVI and those without AV reintervention. Outcomes, including all-cause mortality and a composite endpoint defined as all-cause mortality and heart failure rehospitalization, were assessed using the Kaplan-Meier method.
RESULTS
A total of 46,777 patients who underwent TAVI were included in the analysis. The median age was 81 (IQ range: 78, 85) years, and 42% were male. Among these, 501 patients (1.1%) required AV reintervention during a median follow-up of 2.4 (IQ range: 1.1, 4.2) years. Specifically, 247 (49%) underwent re-SAVR, 224 (45%) underwent re-TAVI, and 30 underwent re-BAV (6%). The median time to AV reintervention differed significantly: 0.7 years for re-SAVR, 3.8 years for re-TAVI, and 0.5 years for re-BAV (p<0.001). Endocarditis was the leading diagnosis for AV reintervention in the majority of patients undergoing re-SAVR (67%). In contrast, aortic stenosis was predominant (68%) among those undergoing re-TAVI and aortic regurgitation was the most common diagnosis (67%) among patients undergoing re-BAV.
According to Kaplan-Meier analysis, over a 3-year follow-up the rate of all-cause mortality as well as the composite endpoint was highest among patients undergoing re-SAVR (p<0.001 versus re-TAVI patients; p=0.28 and 0.44 versus re-BAV, respectively, see Fig. 1). In a 3:1 propensity score-matched comparison, 1-year outcomes for re-TAVI patients were similar to TAVI patients without AV reintervention with respect to all-cause mortality rates and the composite endpoint (p=0.40 for all-cause mortality and p=0.10 for composite endpoint, respectively see Fig. 2). However, by the 3-year follow-up, re-TAVI patients had significantly worse outcome compared to TAVI patients without AV reintervention (p=0.04 for all-cause mortality and p=0.003 for composite endpoint, respectively; see Fig. 2)
CONCLUSION
Based on observational health care data AV reintervention is rarely performed among patients undergoing TAVI. Endocarditis is the most common diagnosis leading to re-SAVR within the first 9 months after index TAVI and those patients have worse outcome compared to re-TAVI patients. In contrast, re-TAVI is mostly performed for late valvular dysfunction after index TAVI with comparable 1-year but inferior 3-year outcomes compared to a matched TAVI patient cohort without AV reintervention.