Predictors of All-Cause Mortality After Successful Transcatheter Aortic Valve Implantation in Patients With Atrial Fibrillation

•Predictors of mortality after aortic valve implantation were identified.•A novel mortality risk assessment tool was developed.•Improved discriminatory ability compared with other risk score models.•Clinical characteristics were associated with greater all-cause mortality. Prevalent and incident atrial fibrillation are common in patients who undergo transcatheter aortic valve implantation and are associated with impaired postprocedural outcomes, including mortality. We determined predictors of long-term mortality in patients with atrial fibrillation after successful transcatheter aortic valve implantation. The EdoxabaN Versus standard of care and theIr effectS on clinical outcomes in pAtients havinG undergonE Transcatheter Aortic Valve Implantation–Atrial Fibrillation (ENVISAGE-TAVI AF) trial (NCT02943785) was a multicenter, prospective, randomized controlled trial in patients with prevalent or incident atrial fibrillation after successful transcatheter aortic valve implantation who received edoxaban or vitamin K antagonists. A Cox proportional hazard model was performed to identify predictors of all-cause mortality using a stepwise approach for multiple regression analysis. In addition, we assessed the performance of different risk scores and prediction models using ENVISAGE-TAVI AF data. Of 1,426 patients in ENVISAGE-TAVI AF, 178 (12.5%) died during the follow-up period (median 548 days). Our stepwise approach identified greater risk of mortality with older age, impaired renal function, nonparoxysmal atrial fibrillation, excessive alcohol use, New York Heart Association heart failure class III/IV, peripheral artery disease, and history of major bleeding or predisposition to bleeding. The present model (concordance statistic c-statistic 0.67) was a better discriminator than were other frequently used risk scores, such as the Society of Thoracic Surgeons score (c-statistic 0.56); Congestive heart failure, Hypertension, Age ≥75, Diabetes, Stroke, Vascular disease, Age 65 to 74 years, and Sex category (CHA2DS2-VASc) score (c-statistic 0.54); or Hypertension, Abnormal renal/liver function, Stroke, Bleeding history or predisposition, Labile international normalized ratio, Elderly, and Drugs/alcohol concomitantly (HAS-BLED) score (c-statistic 0.58). In ENVISAGE-TAVI AF, several modifiable and nonmodifiable clinical characteristics were significantly associated with greater long-term all-cause mortality. Improved risk stratification to estimate the probability of mortality after successful transcatheter aortic valve implantation in patients with atrial fibrillation may improve long-term patient prognosis. Prevalent and incident atrial fibrillation are common in patients who undergo transcatheter aortic valve implantation and are associated with impaired postprocedural outcomes, including mortality. We determined predictors of long-term mortality in patients with atrial fibrillation after successful transcatheter aortic valve implantation. The EdoxabaN Versus standard of care and theIr effectS on clinical outcomes in pAtients havinG undergonE Transcatheter Aortic Valve Implantation–Atrial Fibrillation (ENVISAGE-TAVI AF) trial (NCT02943785) was a multicenter, prospective, randomized controlled trial in patients with prevalent or incident atrial fibrillation after successful transcatheter aortic valve implantation who received edoxaban or vitamin K antagonists. A Cox proportional hazard model was performed to identify predictors of all-cause mortality using a stepwise approach for multiple regression analysis. In addition, we assessed the performance of different risk scores and prediction models using ENVISAGE-TAVI AF data. Of 1,426 patients in ENVISAGE-TAVI AF, 178 (12.5%) died during the follow-up period (median 548 days). Our stepwise approach identified greater risk of mortality with older age, impaired renal function, nonparoxysmal atrial fibrillation, excessive alcohol use, New York Heart Association heart failure class III/IV, peripheral artery disease, and history of major bleeding or predisposition to bleeding. The present model (concordance statistic c-statistic 0.67) was a better discriminator than were other frequently used risk scores, such as the Society of Thoracic Surgeons score (c-statistic 0.56); Congestive heart failure, Hypertension, Age ≥75, Diabetes, Stroke, Vascular disease, Age 65 to 74 years, and Sex category (CHA2DS2-VASc) score (c-statistic 0.54); or Hypertension, Abnormal renal/liver function, Stroke, Bleeding history or predisposition, Labile international normalized ratio, Elderly, and Drugs/alcohol concomitantly (HAS-BLED) score (c-statistic 0.58). In ENVISAGE-TAVI AF, several modifiable and nonmodifiable clinical characteristics were significantly associated with greater long-term all-cause mortality. Improved risk stratification to estimate the probability of mortality after successful transcatheter aortic valve implantation in patients with atrial fibrillation may improve long-term patient prognosis. Over the last 2 decades, transcatheter aortic valve implantation (TAVI) has become a widely adopted alternative to surgical aortic valve replacement in patients with severe symptomatic aortic stenosis.1Otto CM Nishimura RA Bonow RO Carabello BA Erwin 3rd, JP Gentile F Jneid H Krieger EV Mack M McLeod C O'Gara PT Rigolin VH Sundt 3rd, TM Thompson A Toly C 2020 ACC/AHA guideline for the management of patients with valvular heart disease: executive summary: a report of the American College of Cardiology/American Heart Association Joint Committee on clinical practice guidelines.Circulation. 2021; 143: e35-e71PubMed Google Scholar,2Vahanian A Beyersdorf F Praz F Milojevic M Baldus S Bauersachs J Capodanno D Conradi L De Bonis M De Paulis R Delgado V Freemantle N Gilard M Haugaa KH Jeppsson A Jüni P Pierard L Prendergast BD Sádaba JR Tribouilloy C Wojakowski W ESC/EACTS Scientific Document Group2021 ESC/EACTS guidelines for the management of valvular heart disease: developed by the task force for the management of valvular heart disease of the European Society of Cardiology (ESC) and the European Association for Cardio-Thoracic Surgery (EACTS).Rev Esp Cardiol (Engl Ed). 2022; 75: 524PubMed Google Scholar Advances in bioprosthetic valve technologies and transcatheter delivery techniques, coupled with increased operator experience, have facilitated the expansion of TAVI indications from patients at high risk to those at moderate and low risk. Despite this progress, short- and long-term complications after TAVI occur and may affect prognosis.3Winter MP Bartko P Hofer F Zbiral M Burger A Ghanim B Kastner J Lang IM Mascherbauer J Hengstenberg C Goliasch G Evolution of outcome and complications in TAVR: a meta-analysis of observational and randomized studies.Sci Rep. 2020; 10: 15568Crossref PubMed Scopus (48) Google Scholar After TAVI, approximately 1/3 of patients present with atrial fibrillation (AF); the minority of these patients present with new-onset AF. Previous studies indicated that AF is a well-established risk factor for mortality after successful TAVI.4Tarantini G Mojoli M Windecker S Wendler O Lefèvre T Saia F Walther T Rubino P Bartorelli AL Napodano M D'Onofrio A Gerosa G Iliceto S Vahanian A Prevalence and impact of atrial fibrillation in patients with severe aortic stenosis undergoing transcatheter aortic valve replacement: an analysis from the SOURCE XT prospective multicenter registry.JACC Cardiovasc Interv. 2016; 9: 937-946Crossref PubMed Scopus (133) Google Scholar, 5Mojoli M Gersh BJ Barioli A Masiero G Tellaroli P D'Amico G Tarantini G Impact of atrial fibrillation on outcomes of patients treated by transcatheter aortic valve implantation: a systematic review and meta-analysis.Am Heart J. 2017; 192: 64-75Crossref PubMed Scopus (46) Google Scholar, 6Guedeney P Chieffo A Snyder C Mehilli J Petronio AS Claessen BE Sartori S Lefèvre T Presbitero P Capranzano P Tchétché D Iadanza A Sardella G Van Mieghem NM Chandrasekhar J Vogel B Sorrentino S Kalkman DN Meliga E Dumonteil N Fraccaro C Trabattoni D Mikhail G Ferrer MC Naber C Kievit P Baber U Sharma S Morice MC Mehran R WIN-TAVI InvestigatorsImpact of baseline atrial fibrillation on outcomes among women who underwent contemporary transcatheter aortic valve implantation (from the Win-TAVI registry).Am J Cardiol. 2018; 122: 1909-1916Abstract Full Text Full Text PDF PubMed Scopus (17) Google Scholar However, other predictors of mortality in patients with AF who undergo TAVI remain uncertain. A meta-analysis of 4 randomized trials investigated causes of mortality in patients with AF who received non–vitamin K antagonist (VKA) oral anticoagulants (NOACs) or warfarin for prevention of stroke and systemic embolism.7Gómez-Outes A Lagunar-Ruíz J Terleira-Fernández AI Calvo-Rojas G Suárez-Gea ML Vargas-Castrillón E Causes of death in anticoagulated patients with atrial fibrillation.J Am Coll Cardiol. 2016; 68: 2508-2521Crossref PubMed Scopus (194) Google Scholar Older age, low creatinine clearance (CrCl), heart failure, diabetes, and male sex were significantly associated with a greater risk of mortality in patients with AF who were on anticoagulants.7Gómez-Outes A Lagunar-Ruíz J Terleira-Fernández AI Calvo-Rojas G Suárez-Gea ML Vargas-Castrillón E Causes of death in anticoagulated patients with atrial fibrillation.J Am Coll Cardiol. 2016; 68: 2508-2521Crossref PubMed Scopus (194) Google Scholar Notably, this meta-analysis included a broad cohort of patients with AF and was not specific to patients with AF who undergo A Lagunar-Ruíz J Terleira-Fernández AI Calvo-Rojas G Suárez-Gea ML Vargas-Castrillón E Causes of death in anticoagulated patients with atrial fibrillation.J Am Coll Cardiol. 2016; 68: 2508-2521Crossref PubMed Scopus (194) Google Scholar stratification are in the probability of complications in patients with AF and to the and of undergoing In the of a well-established risk score for TAVI, the Society of Thoracic Surgeons score is used in the Heart of patients with symptomatic aortic However, this score was from of patients who underwent surgical aortic valve replacement and may not estimate the risk in for have several to risk was by ability to different from the from the scores were F Capodanno D Tarantini G F C ML N M Napodano M P Saia F G C V P A Presbitero P F E P F S G S M F D'Amico M C and of the TAVI score for after transcatheter aortic valve implantation for aortic J Cardiol. Full Text Full Text PDF PubMed Google Scholar, B C D H K P J P A M A E M Vahanian A Gilard M 2 of mortality after transcatheter aortic valve implantation: risk assessment using a PubMed Scopus Google Scholar, J J and mortality after transcatheter aortic valve Am Coll Cardiol. 2016; 68: PubMed Scopus Google Scholar, P T A J F T L D B D P A L risk model for outcome after transcatheter aortic valve J Cardiol. 2020; Full Text Full Text PDF PubMed Scopus Google Scholar, T and mortality risk after transcatheter aortic valve Cardiovasc 2021; PubMed Scopus Google Scholar The EdoxabaN Versus standard of care and theIr effectS on clinical outcomes in pAtients havinG undergonE Transcatheter Aortic Valve Implantation–Atrial Fibrillation (ENVISAGE-TAVI trial was a prospective, multicenter, trial that compared the and of a factor with those of in patients with AF after successful Mieghem NM M Hengstenberg C H Mehran R D L R P H Lang F M K R F P J E J P Capranzano P T R Baber U A P H C M R S ENVISAGE-TAVI AF vitamin K antagonist for atrial fibrillation after J 2021; PubMed Scopus Google Mieghem NM M M Mehran R E Baber U Hengstenberg C M C S R P standard of care and on clinical outcomes in patients transcatheter aortic valve implantation in atrial and of the ENVISAGE-TAVI AF Heart J. 2018; PubMed Scopus Google Scholar The of all-cause mortality was for patients treated with edoxaban and for those treated with Mieghem NM M Hengstenberg C H Mehran R D L R P H Lang F M K R F P J E J P Capranzano P T R Baber U A P H C M R S ENVISAGE-TAVI AF vitamin K antagonist for atrial fibrillation after J 2021; PubMed Scopus Google Scholar is to the risk of death in this and to to mortality of ENVISAGE-TAVI AF to identify the predictors of all-cause mortality after successful TAVI in patients with an for oral to prevalent or incident AF. The of the ENVISAGE-TAVI AF trial is in Mieghem NM M M Mehran R E Baber U Hengstenberg C M C S R P standard of care and on clinical outcomes in patients transcatheter aortic valve implantation in atrial and of the ENVISAGE-TAVI AF Heart J. 2018; PubMed Scopus Google Scholar The trial was in with the of on and and to clinical The trial was by the and for patients patients were with prevalent or incident AF after successful with increased bleeding and to edoxaban and were patient were from were to edoxaban or a and after The edoxaban was to of the were to and of Mieghem NM M Hengstenberg C H Mehran R D L R P H Lang F M K R F P J E J P Capranzano P T R Baber U A P H C M R S ENVISAGE-TAVI AF vitamin K antagonist for atrial fibrillation after J 2021; PubMed Scopus Google Scholar patients treated with the international normalized was to to to for patients in oral was at the of the of the was during the period was alternative anticoagulants other than the by for used as a or the as was on for such as for after and with other such as TAVI with or during the period were to or modifiable and nonmodifiable were for with all-cause mortality after included age, of AF or major bleeding or predisposition to bleeding New York Heart Association heart failure class of with than of heart or class of heart failure at G J S Mack and outcomes after transcatheter valve in heart the Cardiovasc Interv. 2020; PubMed Scopus Google Scholar excessive alcohol or use, or of with of renal or creatinine or and international normalized international normalized in In to these of for of oral as and and were were performed in the included patients randomized of received a of the The period for the analysis was the from to the last the at an or and baseline characteristics were using and compared patients with and mortality during the were using the or on the of the hazard regression were performed to the hazard of for all-cause mortality after A Cox proportional hazard model was performed to identify predictors of all-cause mortality using a stepwise approach for the multiple regression analysis. A of was for of a in the model and for a to remain in the are as hazard with for the model with were we assessed the performance of risk scores and prediction models using the ENVISAGE-TAVI AF data. included the Congestive heart failure, Hypertension, Age ≥75, Diabetes, Stroke, Vascular disease, Age 65 to 74 years, and Sex category (CHA2DS2-VASc) Hypertension, Abnormal renal/liver function, Stroke, Bleeding history or predisposition, Labile international normalized ratio, Elderly, and Drugs/alcohol concomitantly (HAS-BLED) Aortic and B C D H K P J P A M A E M Vahanian A Gilard M 2 of mortality after transcatheter aortic valve implantation: risk assessment using a PubMed Scopus Google Scholar of and of for the of Aortic D M C P M G F F C Capranzano P S F risk tool for prediction of mortality after transcatheter aortic valve J Cardiol. Full Text Full Text PDF PubMed Scopus Google Scholar and Aortic Valve W L T A F A J G The aortic valve score J 2017; PubMed Scopus Google Scholar The mortality risk score was from the present stepwise The from the Cox proportional hazard model were by and to the to for model A risk score was on the of the model and were and The and of patients who died and for risk category were The mortality was as were performed using Of the 1,426 patients with AF after successful TAVI in ENVISAGE-TAVI AF, a of 178 patients (12.5%) died during the trial follow-up period 548 days). The baseline characteristics for patients who died and those who are in The patients who died were older than those who standard years, with a greater of artery disease and disease and and In addition, patients who died were than those who to have a history of or score was the 2 the of of patients who died and of patients who of the for edoxaban to baseline patient and clinical to artery artery of renal systemic major bleeding or predisposition to for for of the edoxaban included used as an in and with a used as an in AF atrial Congestive heart failure, Hypertension, Age years, Diabetes, disease, Age 65 to 74 years, Sex creatinine Hypertension, Abnormal function, Bleeding history or predisposition, Labile Elderly, international normalized standard Society of Thoracic alcohol are as for of the edoxaban included used as an in and with a used as an in atrial Congestive heart failure, Hypertension, Age years, Diabetes, disease, Age 65 to 74 years, Sex creatinine Hypertension, Abnormal function, Bleeding history or predisposition, Labile Elderly, international normalized standard Society of Thoracic in a are as associated with long-term all-cause mortality were identified by analysis Age history of a systemic history of peripheral artery disease class or for of AF excessive alcohol disease and were significantly associated with all-cause analysis of baseline of to systemic artery class or of of included and of or included and performed major bleeding or predisposition to artery for for of the edoxaban included used as an in and with a used as an in AF atrial creatinine hazard international normalized non–vitamin K antagonist oral New York Heart TAVI transcatheter aortic valve vitamin K alcohol were included and for of the edoxaban included used as an in and with a used as an in atrial creatinine hazard international normalized non–vitamin K antagonist oral New York Heart TAVI transcatheter aortic valve vitamin K in a were regression analysis that the of long-term all-cause mortality were to to of AF to excessive alcohol to class or to history of peripheral artery disease to and major bleeding or predisposition to bleeding to of mortality from multiple regression artery class or major bleeding or predisposition to nonparoxysmal alcohol were atrial creatinine hazard New York Heart in a were AF atrial creatinine hazard New York Heart A prediction model for all-cause mortality was with the clinical identified by multiple regression analysis as predictors of mortality. The is in The statistic for model was to models were on and score The for the present model was by the score model to score model to and the score model to 0.58). The the for the present stepwise model was greater than that of models on the score score score score score and score for these models are in The mortality risk score from the model is in excessive alcohol the by and were on risk to to and mortality by risk score category are in A in mortality was from to and score the of the this in the ENVISAGE-TAVI AF is the to report predictors of long-term mortality after successful TAVI in patients with an for oral to prevalent or incident AF. death in of patients the predictors of long-term all-cause mortality were modifiable excessive alcohol and nonmodifiable age, AF class or peripheral artery disease, and major bleeding or predisposition to risk The or not as an of mortality. of several other risk scores to the present mortality risk The prediction of mortality on the of risk is to clinical the and all-cause mortality risk is not in this cohort of patients with AF after TAVI, the identified may as for increased risk of mortality in these risk models are from surgical aortic valve replacement and are used for mortality risk have ability in TAVI in of such as those with AF. In this on patients who underwent TAVI with an for oral to prevalent or incident AF, several clinical predictors of long-term mortality were age, renal function, heart failure, peripheral artery disease, and bleeding. are with in patients with AF and in patients who undergo P Chieffo A Snyder C Mehilli J Petronio AS Claessen BE Sartori S Lefèvre T Presbitero P Capranzano P Tchétché D Iadanza A Sardella G Van Mieghem NM Chandrasekhar J Vogel B Sorrentino S Kalkman DN Meliga E Dumonteil N Fraccaro C Trabattoni D Mikhail G Ferrer MC Naber C Kievit P Baber U Sharma S Morice MC Mehran R WIN-TAVI InvestigatorsImpact of baseline atrial fibrillation on outcomes among women who underwent contemporary transcatheter aortic valve implantation (from the Win-TAVI registry).Am J Cardiol. 2018; 122: 1909-1916Abstract Full Text Full Text PDF PubMed Scopus (17) Google A Lagunar-Ruíz J Terleira-Fernández AI Calvo-Rojas G Suárez-Gea ML Vargas-Castrillón E Causes of death in anticoagulated patients with atrial fibrillation.J Am Coll Cardiol. 2016; 68: 2508-2521Crossref PubMed Scopus (194) Google Scholar is associated with increased risk of mortality of the of AF and the A E D C O F S S M R C T T P M S D Windecker S S outcomes after transcatheter aortic valve replacement: from the registry.JACC Cardiovasc Interv. 2021; PubMed Scopus Google D S J D F Mack outcomes at transcatheter aortic valve PubMed Scopus Google Scholar heart failure, and peripheral disease are well-established risk for after D S J D F Mack outcomes at transcatheter aortic valve PubMed Scopus Google M T T R A T F N T M F K M S H K A S K risk model for mortality after transcatheter aortic valve Cardiovasc Interv. 2021; PubMed Scopus Google Scholar In the excessive alcohol and mortality after aortic valve has not several studies that alcohol the risk of and S A L G and mortality risk in to alcohol in patients with PubMed Scopus Google Scholar, RA M and the and Am Coll Cardiol. PubMed Scopus Google Scholar, A S M P G P A P A J R S A R V A L W B M S and disease, to and a prospective cohort Full Text Full Text PDF PubMed Scopus Google Scholar excessive alcohol was the modifiable risk factor identified by of alcohol TAVI may postprocedural outcomes and improve long-term prognosis. of mortality identified in and frequently in patients who undergo of is a history of bleeding or bleeding bleeding complications are associated with death after R S E T Claessen BE A F H M H Impact of bleeding on mortality after from a analysis of the of to clinical and and outcomes with and in Cardiovasc Interv. PubMed Scopus Google and impact of bleeding complications in and J Cardiol. Full Text Full Text PDF PubMed Scopus Google Scholar studies a impact of bleeding complications on mortality after P Mack M K R Bleeding complications after surgical aortic valve replacement compared with transcatheter aortic valve replacement: from the trial of aortic transcatheter Am Coll Cardiol. PubMed Scopus Google B E M C A JP A H predictors and impact of bleeding after transcatheter aortic valve implantation using the PubMed Scopus Google Scholar history of bleeding is an of the is frequently used to bleeding risk in patients with R R A novel score (HAS-BLED) to risk of major bleeding in patients with atrial the Heart Full Text Full Text PDF PubMed Scopus Google Scholar may a high bleeding risk as a for mortality in patients with AF who undergo the of AF a impact on mortality after TAVI, with nonparoxysmal AF a of mortality than was with a meta-analysis that was associated with a significantly greater risk of death than was T P The impact of atrial fibrillation on the risk of and a systematic review and Heart J. 2016; PubMed Scopus Google Scholar clinical guidelines oral of AF for management of in patients with atrial a report of the American College of Cardiology Document Am Coll Cardiol. 2017; PubMed Scopus Google Scholar ENVISAGE-TAVI AF included patients with prevalent or incident AF after successful TAVI who were to edoxaban or several studies the of for the prevention of and death in patients with E AL J W M J M M warfarin in patients with atrial J PubMed Scopus Google M J D A MC R G D M Gersh BJ S S J P P P A J L and warfarin in patients with atrial J PubMed Scopus Google Scholar all-cause mortality were indicated for patients treated with edoxaban and in the ENVISAGE-TAVI AF Mieghem NM M Hengstenberg C H Mehran R D L R P H Lang F M K R F P J E J P Capranzano P T R Baber U A P H C M R S ENVISAGE-TAVI AF vitamin K antagonist for atrial fibrillation after J 2021; PubMed Scopus Google Scholar The present the or was not an of all-cause the significantly of major bleeding with edoxaban than with in the trial to Mieghem NM M Hengstenberg C H Mehran R D L R P H Lang F M K R F P J E J P Capranzano P T R Baber U A P H C M R S ENVISAGE-TAVI AF vitamin K antagonist for atrial fibrillation after J 2021; PubMed Scopus Google Scholar A randomized trial of in patients with AF after TAVI mortality JP S A Van E T P E G after valve implantation for aortic the randomized Heart J. 2018; PubMed Scopus Google Scholar a meta-analysis in all-cause mortality patients and those E R RA oral anticoagulants vitamin K in patients with atrial fibrillation after transcatheter aortic valve replacement: a systematic review and Cardiovasc Interv. 2022; PubMed Scopus Google Scholar a risk score for mortality after successful TAVI was from the model in the present risk score is the risk assessment tool in patients long-term oral for AF after a successful model greater ability of mortality than other risk scores and risk prediction models in patients with AF and those in patients who undergo may by in baseline characteristics of the cohort used to the score and the ENVISAGE-TAVI AF the for risk scores, for patients with an for oral to AF. are several of this to patients in the ENVISAGE-TAVI AF were older and at surgical risk. the present may not to a at surgical risk. In addition, the predictors are to patients with an for oral of AF. this analysis on clinical that are assessed by may other characteristics not assessed in this that may affect the of mortality. the of new-onset AF is to for after T A R A atrial fibrillation after transcatheter aortic valve replacement: a systematic review and Cardiovasc Interv. 2022; PubMed Scopus Google Scholar the of or after TAVI are from this in other TAVI including patients an for oral to AF and in other with AF. In patients with AF after successful TAVI, several modifiable alcohol and nonmodifiable class or peripheral artery disease, and major bleeding or predisposition to clinical risk were significantly associated with greater long-term all-cause mortality. A of these risk may for risk stratification of patients with AF who undergo The this on to the is a clinical for and is a clinical for and Hengstenberg is a clinical for and and for for and from is a for and Van Mieghem or from and is an of is a clinical for and from and and from and from and Mehran to for from and and from for from and is an of is an of is an of to for from and and from has to The for to in this was by Van of a with