Long-Term Cause of Death in Patients Who Underwent Transcatheter Aortic Valve Implantation

Abstract

As our knowledge on treatment with transcatheter aortic valve implantation (TAVI) increases and more implantations are conducted, we need knowledge on how TAVI affects the end of life. Long-term causes of death remain sparsely described. The aim of the study was to examine differences in the cause of death according to time from TAVI. All patients who underwent TAVI in Denmark from 2008 to 2017 were matched on gender, age, and calendar year with controls from the background population (1:4). Mortality and the proportion of cardiovascular and noncardiovascular death was assessed at 1-year time points during follow-up. A total of 3,434 patients receiving TAVI and 13,672 controls were identified. The median follow-up was 2.67 years for patients receiving TAVI and 2.90 years for controls. Among patients receiving TAVI, 1,254 deaths (36.5%) were recorded, with 46.7% being from cardiovascular causes. The corresponding numbers for controls were 3,338 deaths (24.4%) and 27.2% being from cardiovascular causes. The proportion of cardiovascular deaths decreased from 53.8% in the first year after TAVI to 32.7% among those who died >7 years from TAVI (p = 0.008 for trend). For controls, no difference was seen in the proportion of cardiovascular death regardless of follow-up time. In conclusion, with data from nationwide registries, we provide results reassuring that patients with long-term survival from TAVI resemble the general public regarding the cause of death. As our knowledge on treatment with transcatheter aortic valve implantation (TAVI) increases and more implantations are conducted, we need knowledge on how TAVI affects the end of life. Long-term causes of death remain sparsely described. The aim of the study was to examine differences in the cause of death according to time from TAVI. All patients who underwent TAVI in Denmark from 2008 to 2017 were matched on gender, age, and calendar year with controls from the background population (1:4). Mortality and the proportion of cardiovascular and noncardiovascular death was assessed at 1-year time points during follow-up. A total of 3,434 patients receiving TAVI and 13,672 controls were identified. The median follow-up was 2.67 years for patients receiving TAVI and 2.90 years for controls. Among patients receiving TAVI, 1,254 deaths (36.5%) were recorded, with 46.7% being from cardiovascular causes. The corresponding numbers for controls were 3,338 deaths (24.4%) and 27.2% being from cardiovascular causes. The proportion of cardiovascular deaths decreased from 53.8% in the first year after TAVI to 32.7% among those who died >7 years from TAVI (p = 0.008 for trend). For controls, no difference was seen in the proportion of cardiovascular death regardless of follow-up time. In conclusion, with data from nationwide registries, we provide results reassuring that patients with long-term survival from TAVI resemble the general public regarding the cause of death. Transcatheter aortic valve implantation (TAVI) is at least noninferior to surgical aortic valve replacement across all surgical risk profiles.1Mack MJ Leon MB Thourani VH Makkar R Kodali SK Russo M Kapadia SR Malaisrie SC Cohen DJ Pibarot P Leipsic J Hahn RT Blanke P Williams MR McCabe JM Brown DL Babaliaros V Goldman S Szeto WY Genereux P Pershad A Pocock SJ Alu MC Webb JG Smith CR PARTNER 3 InvestigatorsTranscatheter aortic-valve replacement with a balloon-expandable valve in low-risk patients.N Engl J Med. 2019; 380: 1695-1705Crossref PubMed Scopus (2780) Google Scholar, 2Leon MB Smith CR Mack MJ Makkar RR Svensson LG Kodali SK Thourani VH Tuzcu EM Miller DC Herrmann HC Doshi D Cohen DJ Pichard AD Kapadia S Dewey T Babaliaros V Szeto WY Williams MR Kereiakes D Zajarias A Greason KL Whisenant BK Hodson RW Moses JW Trento A Brown DL Fearon WF Pibarot P Hahn RT Jaber WA Anderson WN Alu MC Webb JG PARTNER 2 InvestigatorsTranscatheter or surgical aortic-valve replacement in intermediate-risk patients.N Engl J Med. 2016; 374: 1609-1620Crossref PubMed Scopus (3522) Google Scholar, 3Leon MB Smith CR Mack M Miller DC Moses JW Svensson LG Tuzcu EM Webb JG Fontana GP Makkar RR Brown DL Block PC Guyton RA Pichard AD Bavaria JE Herrmann HC Douglas PS Petersen JL Akin JJ Anderson WN Wang D Pocock S. Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery.N Engl J Med. 2010; 363: 1597-1607Crossref PubMed Scopus (5682) Google Scholar, 4Smith CR Leon MB Mack MJ Miller DC Moses JW Svensson LG Tuzcu EM Webb JG Fontana GP Makkar RR Williams M Dewey T Kapadia S Babaliaros V Thourani VH Corso P Pichard AD Bavaria JE Herrmann HC Akin JJ Anderson WN Wang D Pocock SJ PARTNER Trial InvestigatorsTranscatheter versus Surgical Aortic-Valve Replacement in High-Risk Patients.N Engl J Med. 2011; 364: 2187-2198Crossref PubMed Scopus (5005) Google Scholar, 5Thyregod HGH Ihlemann N Jørgensen TH Nissen H Kjeldsen BJ Petursson P Chang Y Franzen OW Engstrøm T Clemmensen P Hansen PB Andersen LW Steinbrüchel DA Olsen PS Søndergaard L. Five-year clinical and echocardiographic outcomes from the Nordic Aortic Valve Intervention (NOTION) randomized clinical trial in patients at lower surgical risk.Circulation. 2019; 139: 2714-2723Crossref PubMed Scopus (176) Google Scholar Short-term mortality and cause of death is well described among patients receiving TAVI6Xiong TY Liao YB Zhao ZG Xu YN Wei X Zuo ZL Li YJ Cao JY Tang H Jilaihawi H Feng Y Chen M. Causes of death following transcatheter aortic valve replacement: a systematic review and meta-analysis.J Am Heart Assoc. 2015; 4e002096Crossref Scopus (42) Google Scholar; however, cause of death with long-term follow-up is sparse and data on this area may help us acknowledge potentially preventable deaths. It has previously been shown that the mortality among patients receiving TAVI is significantly higher within the first 3 months after TAVI compared with matched controls from the background population; however, this associated risk difference diminishes after the first 3 months after TAVI.7Jakobsen L Terkelsen CJ Søndergaard L De Backer O Aarøe J Nissen H Johnsen SP Christiansen EH. Short- and long-term mortality and stroke risk after transcatheter aortic valve implantation.Am J Cardiol. 2018; 121: 78-85Abstract Full Text Full Text PDF PubMed Scopus (15) Google Scholar An extended question to this finding is whether causes of death among patients receiving TAVI will resemble causes of death among controls from a background population when long-term follow-up is studied. Knowledge on TAVI increases rapidly and as more transcatheter implantations are conducted, we need data on how TAVI affects the end of life.8Carroll JD Mack MJ Vemulapalli S Herrmann HC Gleason TG Hanzel G Deeb GM Thourani VH Cohen DJ Desai N Kirtane AJ Fitzgerald S Michaels J Krohn C Masoudi FA Brindis RG Bavaria JE. STS-ACC TVT registry of transcatheter aortic valve replacement.J Am Coll Cardiol. 2020; 76: 2492-2516Crossref PubMed Scopus (391) Google Scholar Our hypothesis is that TAVI may affect the end of life for these patients because cardiovascular causes may represent a smaller proportion.9Food and Drug Administration. FDA approves expanded indication for two transcatheter heart valves for patients at intermediate risk for death or complications associated with open-heart surgery. Available at: https://www.fda.gov/news-events/press-announcements/fda-approves-expanded-indication-two-transcatheter-heart-valves-patients-intermediate-risk-death-or. Accessed on September 18, 2022.Google Scholar,10Food and Drug Administration. FDA expands indication for several transcatheter heart valves to patients at low risk for death or major complications associated with open-heart surgery. Available at: https://www.fda.gov/news-events/press-announcements/fda-expands-indication-several-transcatheter-heart-valves-patients-low-risk-death-or-major. Accessed on September 18, 2022.Google Scholar Thus, we aimed to evaluate and compare the pattern of cause of death in a nationwide cohort of patients treated with TAVI from 2008 to 2017 with a matched cohort from the general population. All Danish citizens are given a unique identifier, which makes it possible to link nationwide registries. For the purpose of this study, 4 registries were linked: (1) the National Patient Registry, (2) the Danish National Population Registry, (3) the National Prescription Registry, and (4) the Cause of Death Registry. The National Patient Registry was initiated in 1977 and provide data on all hospital admissions (inpatient and outpatient visits). Every visit is provided with a primary diagnosis and up to several secondary diagnosis codes given by a physician at discharge, which is mandatory. The codes are based on the International Classifications of Diseases, Tenth Revision (ICD-10). Medical procedures, including surgical procedures, are also given in the National Patient Registry, categorized by the classification of the Nordic Medico-Statistical Committee. The Danish Population Registry provides data on the date of birth, gender, and emigration status. The National Prescription Registry provides data on all prescriptions redeemed from a Danish Pharmacy since 1994. The registries are of high quality and previously described in detail.11Kildemoes HW Sørensen HT Hallas J. The Danish National prescription registry.Scand J Public Health. 2011; 39: 38-41Crossref PubMed Scopus (1689) Google Scholar, 12Schmidt M Schmidt SAJ Sandegaard JL Ehrenstein V Pedersen L Sørensen HT. The Danish National Patient Registry: a review of content, data quality, and research potential.Clin Epidemiol. 2015; 7: 449-490Crossref PubMed Scopus (2644) Google Scholar, 13Schmidt M Pedersen L Sørensen HT. The Danish Civil Registration System as a tool in epidemiology.Eur J Epidemiol. 2014; 29: 541-549Crossref PubMed Google Scholar The Cause of Death Registry will be described in details in a subsequent section. The study population consisted of patients who underwent first-time TAVI in the period of 2008 to 2017, recruited from all 4 TAVI centers in Denmark. We included patients who underwent ministernotomy, transapical, and transfemoral insertions of the prosthetic valve (procedure codes provided in Supplementary Table 1). Since 2015, transfemoral approach has been the method of choice in Denmark. Patients who died during hospitalization for TAVI were excluded. The primary outcome of this study was to compare the proportion of cardiovascular and noncardiovascular causes of death according to time since TAVI and compare these proportions with a gender-, age-, and calendar-matched controls from the background population (a matched cohort resembling the general public). The secondary outcomes were to examine long-term mortality after TAVI and to identify differences in mortality and causes of death according to calendar periods. Patients were followed up from the date of discharge until the date of death, December 31 2018, or a maximum of 9 years of follow-up, whichever came first. Cause of death was assessed from the Cause of Death Registry. The Danish Cause of Death Registry was established in 1970 and is mandatory.14Helweg-Larsen K. The Danish register of causes of death.Scand J Public Health. 2011; 39: 26-29Crossref PubMed Google Scholar The cause of death certificate in Denmark is constructed so that the doctor is obligated to provide an “underlying cause of death” and several following causes of death, describing the continuum of the fatal course. The diagnoses provided in the Cause of Death Registry is based on the ICD-10. We identified the cause of death from the death certificate's “underlying cause of death” and we categorized the cause of death into cardiovascular and noncardiovascular death. A cardiovascular death was defined as any cardiovascular diagnosis code (ICD-10 code “I”). Data on the cause of death was provided up until December 31, 2018, which is why only patients up until December 31 2017 were included, thus allowing at least 1 year of follow-up for every patient unless the patient died during follow-up. Medical history any time before hospitalization for TAVI was assessed from the National Patient Registry as diagnoses from inpatient or outpatient visits. Concomitant pharmacotherapy was assessed from the National Prescription Registry as a redeemed prescription 6 months before hospitalization for TAVI. Baseline characteristics were provided for patients who underwent TAVI and matched controls with categorical variables presented in counts and percentages and continuous variables with a median and 25 and 75 percentiles. Gender-, age-, and calendar-matched controls from the background population were identified using risk-set matching, with 4 controls for every patient receiving TAVI. Hence, patients were required to be alive and not emigrated before the index date. we aimed to compare differences in causes of death patients receiving TAVI and a cohort resembling the general on were not The of deaths and the proportion of cardiovascular and noncardiovascular causes of death were for every year after the date of discharge after TAVI we year >7 to 9 after TAVI because of The was to the difference in the proportions of cardiovascular mortality during the follow-up for the 2 study of cardiovascular and noncardiovascular death were with a maximum of and 9 years of follow-up for the 2 study were as the of by the follow-up time. The were a For patients and controls with follow-up >7 we identified whether was a difference in the proportion of cardiovascular cause of death using the Baseline characteristics mortality was assessed using and the difference was assessed using the We 2 (1) of cardiovascular and noncardiovascular causes of death were assessed for the following calendar 2008 to to to 2015, and to and (2) in an a of the cardiovascular causes of death for patients receiving TAVI and controls were assessed We 3 the cause of death was categorized as cardiovascular any cardiovascular diagnosis code was provided on the death certificate in the continuum of the fatal course. we compared the proportion of and noncardiovascular causes of death according to time since TAVI to only cardiovascular and noncardiovascular causes of for patients with >7 years of follow-up, characteristics were compared study We identified patients who underwent first-time TAVI from 2008 to who died during hospitalization for TAVI were and a total of 3,434 patients were We identified 13,672 matched controls from the general population. were well matched on gender, age, and calendar year of the of was higher among patients who underwent TAVI 1). The of TAVI were as a transfemoral approach and of the patients and controls were included in to in median years period history Heart = = = = = = = TAVI = transcatheter aortic valve in a = = = = = = = TAVI = transcatheter aortic valve The median follow-up time was 2.67 years and 75 to for patients receiving TAVI and 2.90 years and 75 to for controls. Mortality are and in Table 2 an mortality among patients receiving TAVI. Table 2 also mortality at maximum follow-up of and 9 higher mortality among patients receiving TAVI at all time a maximum of 9 years of follow-up, a total of 1,254 deaths (36.5%) were in patients receiving TAVI, with 46.7% being from cardiovascular causes. The corresponding numbers for controls were 3,338 deaths (24.4%) and 27.2% were from cardiovascular causes. In the first year after TAVI, the proportion of cardiovascular deaths was 53.8% within the first = which decreased to 32.7% of mortality after years from TAVI within the follow-up = which was to cardiovascular death (p = 0.008 for 2 For controls, no difference was seen in the proportion of cardiovascular death during follow-up (p = for 2 For patients and controls with follow-up >7 we no difference in the proportion of cardiovascular causes of death for patients receiving TAVI and for controls, = for Table the of cardiovascular and noncardiovascular deaths at a maximum of and 9 years of follow-up. For patients receiving TAVI, the of cardiovascular death the for noncardiovascular death with follow-up time Among controls, a in cardiovascular and noncardiovascular death was seen with follow-up time and for cardiovascular and noncardiovascular death with cardiovascular noncardiovascular of 1 of 3 of of of of of of 9 of = = = TAVI = transcatheter aortic valve in a in cause of death according to time since The the of deaths and the proportion of cardiovascular and noncardiovascular deaths according to follow-up to 1 year of follow-up, 1 to 2 years of follow-up, 2 to 3 years of follow-up, 3 to 4 years of follow-up, 4 to years of follow-up, to 6 years of follow-up, 6 to years of follow-up, >7 to 9 years of follow-up. = cause of = = = = TAVI = transcatheter aortic valve We identified that the of cardiovascular death, with a maximum of 1-year follow-up, decreased from the calendar period 2008 to to to 2017 among patients receiving TAVI with controls, the for cardiovascular death in patients with TAVI were higher for all calendar For noncardiovascular death, no differences was seen study for the 4 calendar of death with a maximum of 1 of follow-up by calendar death, cardiovascular death, death, noncardiovascular death, = = = in a = = = In the of cardiovascular causes of death for patients receiving TAVI and controls, we that the cause of death was heart was more among controls, heart was more among patients receiving TAVI. Heart as a cause of death, was Table We a cardiovascular cause of death was defined as any cardiovascular diagnosis at the death certificate in the fatal only the cause of The of cardiovascular causes of death from to for patients receiving TAVI. For controls, the from to differences in the results from the were In a we the noncardiovascular causes of death in and noncardiovascular death and identified that the proportion of causes of death was according to time from TAVI and within the first year after = of deaths was to 1). Baseline characteristics were compared for study for patients with follow-up >7 years and no major differences were identified from the study population Table nationwide study mortality and causes of death among patients receiving TAVI and matched controls from the general population. The study 3 major we identified a higher mortality in patients receiving TAVI a matched cohort resembling the general public matched on age, gender, and calendar which was of follow-up time. we that more of patients receiving TAVI, who died within the first year of receiving TAVI, died from a cardiovascular cause of death. For patients who >7 years since the TAVI only of deaths was to cardiovascular which was the proportion of cardiovascular deaths among matched controls. the calendar time to we of cardiovascular causes of death among patients receiving TAVI. higher mortality was identified among patients who underwent TAVI a and cohort from the general which was to the higher of in patients who underwent TAVI. Short-term mortality to for a proportion because the mortality for patients who underwent TAVI is We identified that cardiovascular mortality was the cause of death within the first year of follow-up after TAVI. In with the randomized trial Aortic-Valve for Aortic in Patients which the of TAVI in patients with aortic the causes of death were to cardiovascular the causes of death were to noncardiovascular or MB Smith CR Mack M Miller DC Moses JW Svensson LG Tuzcu EM Webb JG Fontana GP Makkar RR Brown DL Block PC Guyton RA Pichard AD Bavaria JE Herrmann HC Douglas PS Petersen JL Akin JJ Anderson WN Wang D Pocock S. Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery.N Engl J Med. 2010; 363: 1597-1607Crossref PubMed Scopus (5682) Google Scholar TAVI with and intermediate-risk surgical the pattern was seen with a high proportion of cardiovascular causes of death within however, this proportion decreased to to at 1-year MB Smith CR Mack MJ Makkar RR Svensson LG Kodali SK Thourani VH Tuzcu EM Miller DC Herrmann HC Doshi D Cohen DJ Pichard AD Kapadia S Dewey T Babaliaros V Szeto WY Williams MR Kereiakes D Zajarias A Greason KL Whisenant BK Hodson RW Moses JW Trento A Brown DL Fearon WF Pibarot P Hahn RT Jaber WA Anderson WN Alu MC Webb JG PARTNER 2 InvestigatorsTranscatheter or surgical aortic-valve replacement in intermediate-risk patients.N Engl J Med. 2016; 374: 1609-1620Crossref PubMed Scopus (3522) Google CR Leon MB Mack MJ Miller DC Moses JW Svensson LG Tuzcu EM Webb JG Fontana GP Makkar RR Williams M Dewey T Kapadia S Babaliaros V Thourani VH Corso P Pichard AD Bavaria JE Herrmann HC Akin JJ Anderson WN Wang D Pocock SJ PARTNER Trial InvestigatorsTranscatheter versus Surgical Aortic-Valve Replacement in High-Risk Patients.N Engl J Med. 2011; 364: 2187-2198Crossref PubMed Scopus (5005) Google Scholar Data from from a with a higher proportion of cardiovascular causes of death within the first 2 years after TAVI, noncardiovascular causes M R D M of long-term cardiovascular versus mortality and in patients transcatheter aortic valve implantation.Am J Cardiol. 2020; Full Text Full Text PDF PubMed Scopus (2) Google Scholar Our are in with data from previously knowledge with a of a of matched controls resembling the general A previously study from including data up until that cardiovascular causes were within the first 3 months after TAVI, noncardiovascular death was after the first 3 M De Backer O Søndergaard L. and cause of death after transcatheter aortic valve replacement as compared to an and background PubMed Scopus Google Scholar Our a however, from our the proportion of cardiovascular deaths to the general population at a time Our results and the knowledge with nationwide data and follow-up until A from 2015, including with data from causes of death in patients receiving TAVI, that within of of deaths were to cardiovascular causes deaths within = TY Liao YB Zhao ZG Xu YN Wei X Zuo ZL Li YJ Cao JY Tang H Jilaihawi H Feng Y Chen M. Causes of death following transcatheter aortic valve replacement: a systematic review and meta-analysis.J Am Heart Assoc. 2015; 4e002096Crossref Scopus (42) Google Scholar of this proportion was deaths = TY Liao YB Zhao ZG Xu YN Wei X Zuo ZL Li YJ Cao JY Tang H Jilaihawi H Feng Y Chen M. Causes of death following transcatheter aortic valve replacement: a systematic review and meta-analysis.J Am Heart Assoc. 2015; 4e002096Crossref Scopus (42) Google Scholar study included several with which the and be as a our study provided data from a nationwide TY Liao YB Zhao ZG Xu YN Wei X Zuo ZL Li YJ Cao JY Tang H Jilaihawi H Feng Y Chen M. Causes of death following transcatheter aortic valve replacement: a systematic review and meta-analysis.J Am Heart Assoc. 2015; 4e002096Crossref Scopus (42) Google Scholar In a we identified that the 1-year of cardiovascular mortality decreased according to the calendar the 1-year of noncardiovascular mortality according to the calendar periods. finding the and from the of the TAVI the of patients who underwent TAVI has and a population with in years in JE C C C L Søndergaard L in of transcatheter aortic valve replacement and patient a nationwide Heart J. PubMed Scopus Google Scholar We identified that the of patients from cardiovascular causes among patients receiving TAVI was from heart and from heart It is that the proportion of patients from heart be high because several risk for aortic and heart are The proportion of patients from heart may be to in because echocardiographic data were not it been of to examine the causes of death among patients with aortic stenosis not who underwent however, data of are not provided from the registries study has data on causes of death were with up to 1-year follow-up until 2018, and since 2018, the of TAVI has been and more data an JD Mack MJ Vemulapalli S Herrmann HC Gleason TG Hanzel G Deeb GM Thourani VH Cohen DJ Desai N Kirtane AJ Fitzgerald S Michaels J Krohn C Masoudi FA Brindis RG Bavaria JE. STS-ACC TVT registry of transcatheter aortic valve replacement.J Am Coll Cardiol. 2020; 76: 2492-2516Crossref PubMed Scopus (391) Google JE C C C L Søndergaard L in of transcatheter aortic valve replacement and patient a nationwide Heart J. PubMed Scopus Google Scholar Hence, long-term data are provided from patients at low surgical risk are TAVI, and before 2018, TAVI was not for this patient population. may a in patient which may into a difference in the causes of death. of the International Classifications of codes in the Causes of Death Registry are not which is why we described the cause of death in to across more clinical as across the transcatheter aortic and of valve or the study population in more nationwide study the mortality and causes of death among patients receiving TAVI and a matched cohort resembling the general public matched on age, gender, and calendar mortality was higher for patients receiving TAVI the matched controls. We that of patients receiving TAVI, who died within the first year of receiving TAVI, died from a cardiovascular cause of death. For patients who >7 years since the TAVI only of deaths were to cardiovascular which was to the proportion of cardiovascular deaths among matched controls. from and to this research for heart from the The no of to with Supplementary in cause of death according to time since with

Topics & Concepts