Survival after aortic root replacement with a stentless xenograft is determined by patient characteristics

,


CENTRAL MESSAGE
Survival after Freestyle (Medtronic Inc, Minneapolis, Minn) bioprosthesis aortic root replacement resembles the general population for low-risk cases; outcome is thus explained by patient-and disease-related factors rather than the prosthesis per se.
The Freestyle bioprosthesis (Medtronic Inc, Minneapolis, Minn) is a stentless porcine aortic root xenograft prosthesis that can be used in patients with a range of severe aortic root diseases.][5][6] However, the existing literature primarily regards selected populations from highly specialized centers.Hence, Freestyle performance in the most common setting of small-to-medium sized centers with unselected patients has not been determined.We therefore conducted a study including 6 North Atlantic institutions.The objective was to examine clinical outcomes, including overall survival and reinterventions, after Freestyle aortic root replacement, stratified according to indication.

METHODS Study Design and Patients
This observational study included Freestyle aortic root replacements from 1999 to 2018 at 6 institutions: Rigshospitalet, University Hospital of Copenhagen (Copenhagen), Denmark; Karolinska University Hospital (Stockholm), Sweden; Haukeland University Hospital (Bergen), Norway; Link€ oping University Hospital (Link€ oping), Sweden; Montreal Heart Institute (Montreal), Canada; and Landsp ıtali University Hospital of Iceland (Reykjavik), Iceland.The participating centers were sole providers of cardiac surgery in each health region.The inclusion criterion was full root implantation of a Freestyle prosthesis in the aortic position.Patients were identified in local registries and those lost to follow-up within 30 days were excluded, thus including patients who died or had a reintervention before 30 days.Medical records were reviewed for baseline and perioperative data.The study was conducted in accordance with the Declaration of Helsinki, and ethical approval was obtained for all sites (Appendix E1).

Surgical Indications and Technique
We categorized surgical indications to reflect differences in patient characteristics and operative complexity.Primarily elective categories were: (1) aortic root aneurysm (aneurysm), (2) aortic valve replacement in patients with a small aortic annulus and risk of prosthesis-patient mismatch (small root), (3) predominantly reoperative aortic valve surgery or bail-out solution after failed attempted aortic root or valve procedures, both a priori characterized by long cross-clamp time and technical complexity (other).Urgent categories were: (4) native or prosthetic (PVE) valve infective endocarditis, aortic root abscess formation, or a combination (endocarditis, shown in Video 1); and (5) Stanford type A aortic dissection requiring aortic root replacement (type A dissection).The Freestyle was implanted VIDEO 1.This is a video of a redo aortic root replacement in a 76-yearold woman who received an aortic valve bioprosthesis 2 months previously because of aortic valve stenosis.The postoperative course was complicated by mediastinitis, and she has now presented with prosthetic valve endocarditis including vegetations on the valve and in the left ventricular outflow tract.The video shows the explantation of the bioprosthesis and the resection of infected tissue.It then moves on to the reconstruction of the resulting defect in the left ventricular outflow tract and left atrial roof with a bovine pericardial patch.This is followed by the implantation of a full root Freestyle (Medtronic Inc, Minneapolis, Minn) bioprosthesis including the fashioning and reimplantation of the coronary buttons (the coronary ostia, with a cuff of aortic tissue).The case exemplifies one of the complex endocarditis cases, which are included in the endocarditis group in this article.Video available at: https://www.jtcvs.org/article/S0022-5223(21)  01042-4/fulltext.Scanning this QR code will take you to the table of contents to access supplementary information.
The Journal of Thoracic and Cardiovascular Surgery c Volume 164, Number 6 using a modified Bentall approach with the proximal suture line technique and orientation performed at the surgeon's discretion.Running sutures were used for the coronary buttons and also for the anastomosis distal to the native ascending aorta or to a vascular prosthesis.

Institutional Characteristics and Practice
The participating institutions have a yearly volume varying between institutions and over time from 200 to approximately 2000 adult cardiac surgery cases per year and 2-10 surgeons per institution, giving a case load per surgeon ranging from 50 to 300 per year.In all centers, aortic root replacement might be performed by any on-staff surgeon, but complex cases as well as elective aortic root surgery have to a varying degree been concentrated on fewer hands.All participating institutions had the possibility of using a homograft, but most lacked in-house supply throughout the study period and thus generally needed to order a homograft in advance for the specific case.Montreal had access to homografts at all times, Bergen had a local homograft storage up to 2009, and Copenhagen keeps a small storage of only a few grafts in different sizes.The choice ultimately falls on the surgeon.The strategy for postoperative anticoagulation and antiplatelet therapy is similar across institutions; in recent times biological prostheses including the Freestyle patients are only given acetylsalicylic acid but in the first part of the study most institutions gave 2-3 months of warfarin treatment.None of the institutions has differentiated the Freestyle from other biological prostheses in their anticoagulant or antiplatelet regimens.

Outcomes
Early complications were reoperation for bleeding or tamponade, perioperative myocardial infarction as defined in Valve Academic Research Consortium-2, postoperative cerebrovascular events, and permanent pacemaker implantation within 30 days from surgery (see definitions in the Appendix E1). 7 Survival status was extracted from civil registries, which guarantee complete and accurate status for domestic residents, except for very recent deaths, which might be registered with a few weeks' delay.Reinterventions were found according to medical record review up until 2019 and divided into those that were secondary to prosthetic valve failure (valve-related) or those caused by coronary artery pathology (coronary).Pseudoaneurysm formation in relation to the Freestyle bioprosthesis was considered valve-related.Because Copenhagen is a national referral site for complex cases, complete Copenhagen reintervention data was ensured using the civil registry number to cross-link with the Danish National Patient Registry in addition to the medical record review.Patients from Copenhagen or Montreal who were also included in a cross-sectional analysis with cardiac computed tomography (CT) were followed up to the date of the study CT imaging, thus not including findings on the CT imaging or ensuing clinical events, all of which have been previously reported. 8

Statistical Analysis
Continuous variables are presented as median with interquartile range, and categorical variables are presented as percentages and frequencies.Body mass index (BMI) is presented as categorical and continuous data.Patients lost to follow-up were censored at the time of their last clinical visit.Kaplan-Meier estimated survival curves were truncated when a minimum of 10% of the group remained at risk and compared using a log rank test.Significant time interaction between indication and time compromised the proportional hazards assumption.We therefore performed a post hoc landmark analysis (ie, survival of all patients up to 90 days after surgery, and for 90 day-survivors to end of follow-up).Cumulative risk of reintervention was visualized using the Aalen-Johansen estimator with death as competing risk.Hazard ratios and 95% confidence intervals (CIs) were estimated using Cox regression, and follow-up time was truncated when all groups had at least 10% remaining at risk.Results were stratified according to indication.Adjusted analysis of the full cohort was enabled using multiple imputation (10 data sets) of missing values (<5%) using fully conditional specification and chained equations (Stef van Buuren, 2020; mice: multivariate imputation by chained equations.R package version 3.9.0)with the use of Rubin's rule, and conditioned on event, event times, and variables known to influence survival and durability: indication, age, center for surgery, sex, preoperative significant coronary artery disease, presence of atrial fibrillation, and estimated glomerular filtration rate.The Danish general population was chosen as a representative reference for the entire study population; survival was obtained from the Human Mortality Database, and standardized for sex, age, and date of birth through the R package CuRe (Lasse Hjort Jakobsen, 2020; CuRe: parametric cure model estimation.R package version 1.0.0.), and compared with study data using the log rank test.Subgroup analysis of survival was performed per sex.All analyses were performed with R version 3.6.1 (R Foundation for Statistical Computing, Vienna, Austria).

Overall Survival
Figure 1 shows Kaplan-Meier survival curves and landmark analyses.Thirty-day mortality rates were 2.9% for aneurysm, 9.3% for small root, 11.2% for other, 14.6% for endocarditis, and 27.4% for type A dissection (Table 3).Survival of the standardized Danish general population overlapped the intermediate-term survival for 90day survivors of the elective indications (aneurysm, small root, and other), whereas patients with endocarditis and type A dissection had excess mortality, also for 90-day survivors (Figures 1 and E2).Subgroup-analysis of survival according to sex showed similar trends per indication (Figure E3).

Early Complications
Perioperative myocardial infarction occurred in 11 (13.3%)patients with small roots; other groups ranged from 3.2% to 9.0% (Table 4).Four patients had their prosthesis replaced early because of continued bleeding (n ¼ 1) or early suspicion of recurrent endocarditis (n ¼ 3).

DISCUSSION
In this report we describe the outcomes in to our knowledge, the largest published Freestyle aortic root replacement cohort, so far.It testifies to the wide range of aortic root pathologies for which this stentless bioprosthesis is used.Furthermore, because survival of aneurysm patients was similar to the general population, outcome seems determined by patient, more than by prosthesis, characteristics.A summary of the study is shown Figure 3.0][11][12][13][14] In previously reported mixed aortic root replacement cohorts, nonelective surgery accounted for 4%-28% of included cases compared with 46% in the current study, and reoperative surgery accounted for 4%-27% of cases compared with 34% in our study. 2,9-12,15-32Also, complex endocarditis was our second most common indication, and 70% of these patients had PVE.Last, in all groups except aneurysm, the most prevalent New York Heart Association functional classification was III-IV, which is strongly associated with mortality and morbidity after cardiac surgery. 9,14,19The overall high level of disease severity in the current study is further reflected by the high proportion of young patients who received a bioprosthesis.This suggests that the surgeons prioritized operative success, and confidence to the Freestyle as a surgical solution, over long-term durability.A summary of publications for the purpose of comparison of our results with suitable study populations is presented in Table 5. 2,[9][10][11][12][15][16][17][18][19][20][21][22][23][24][25][26][27][28][29][30][31][32]

Survival
[12][15][16][17][18][19][20][21][22][23][24][25][26][27][28][29][30][31][32] Our data compare well with the literature; early and late mortality varies significantly with indication for surgery, with 30-day mortality ranging from 3% for aneurysms to 27% for type A dissections.In all studies, including the present, high rates of mortality occur primarily among patients with type A dissection and those with complex endocarditis.The rather high mortality rate in these groups might be explained by the extended surgical aortic root injury in aortic dissection patients and with invasive root abscess formation or redo setting in patients with endocarditis.2]33 Still, the small root group had a 30-day mortality of 9.3% (n ¼ 8), which is higher than the 4.4% reported in a recent meta-analysis of patients with small aortic roots. 2 Only 1 case was related to an early revascularization, but 4 had perioperative myocardial infarctions.Four of the 8 deceased patients were emergent or urgent cases and the median European System for Cardiac Operative Risk  3 and E1).Freedom from reintervention was estimated by Aalen-Johansen analysis with death as competing risk (reported in Figure E4) stratified for indication for root replacement; curves are truncated at 10% remaining at risk in each indication group, vertical ticks represent censor points, and the 95% confidence interval (CI) is visualized as shading.A, Freedom from valve-related reinterventions including valve dysfunction, pseudoaneurysms, endocarditis, and thromboses requiring reintervention.B, Freedom from coronary reinterventions including percutaneous interventions and coronary artery bypass grafting.C, Crude and adjusted hazard ratios (HRs) obtained using Cox proportional hazards regression.Adjusted according to age at surgery, center for surgery, ischemic heart disease, atrial fibrillation (a-fib), estimated glomerular filtration rate (eGFR), valve size, and suture technique.
the Danish general population.This is contrary to the survival in patients after conventional aortic valve replacement, which have been shown to exhibit a small but significant late excess mortality, especially in this younger age group, compared with the general population. 34This is especially reassuring in light of the previously reported high rates of pseudoaneurysm formation. 8,35However, our data seem to support previous reports of continued excess mortality for endocarditis patients and implies that this could also apply to patients with type A dissection (Figure 1, B, Table 5, and Figure E2, F and L). 13,36On this note, late survival for type A dissection appears to be lower than previously reported for root-involving type A dissection, but because our patients were 6-9 years older at the time of surgery than the patients in studies representing comparably complex cases (Table 5), the true difference is probably small.Still, it cannot be ruled out that pseudoaneurysm formation and rupture is a possible mechanism for this excess mortality, although an association between pseudoaneurysms and

ADULT
indication for Freestyle implantation has not been identified. 8A subgroup analysis showed no marked difference between sexes but was underpowered to refute such a difference with certainty.

Reinterventions
Reintervention rates compare well with previous reports of aortic root replacement for similar indications, regardless of the type of prosthesis used (Table 5).Most of the valverelated reinterventions were because of recurrent endocarditis in the group of endocarditis, again likely reflecting the underlying pathophysiology and comorbidity rather than the choice of prosthesis.Overall freedom from valve-related reintervention during the first 8 years was generally high (94.4%).Reinterventions because of structural valve deterioration were rare.Coronary reinterventions primarily occurred among small root patients, who had a 5-year freedom of 90% (Figure 2, B).Of note, 4 of 7 revascularizations in this group were perioperative and represented 4 of the 7 early revascularizations in the entire study unit.Thus, the small root appears particularly prone to coronary reimplantation complications such as, for example, a kink of the proximal coronary artery, but all early revascularizations must be considered likely complications to coronary artery reimplantation.Even so, all but 1 of the four small root patients and 2 of the 7 patients in all, survived this complication.In all, the authors consider the perioperative risk of kink or other ostial obstruction in relation to aortic root replacement admissible, but the results do underline the importance of coronary positioning when reimplanting the coronary buttons, especially under the narrow conditions seen, for example, in the small root setting.Most of the late coronary lesions were proximal and might represent a complication to the coronary reimplantation.

Early Complications
Overall complication frequencies are shown in Table 4; they reflect patient characteristics, disease severity, and surgical complexity.3][24][26][27][28][29]31,32,37 However, the number of cerebrovascular events in the type A dissection and endocarditis groups were 22.4% and 10.2%, respectively, compared with up to 10% observed in other studies of these indications. 27,31,32Some studies included only permanent stroke whereas others did not specify inclusion (eg, transient ischemic attacks or stroke without sequelae).The current study included all cerebrovascular events whether they were due to the underlying disease, or real intra-or postoperative events, perhaps explaining the higher rate.The high rate of perioperative myocardial and later increased risk of revascularization (hazard ratio, 12.01; 95% CI, 1.99-72.39).However, the perioperative myocardial infarction and the late revascularizations might represent a particular, technical difficulty of coronary reimplantation among the small root patients.

Strengths and Limitations
The multicenter design, large study population, and complete follow-up were strengths of the study.Although our study was limited by lack of cause of death, the comparison of the survival of our cohort with that of the standardized general population allows interpretation of the prosthesis implantation's influence on mortality.The main limitations, additional to those inherent to the retrospective study design, included the lack of detailed serial follow-up echocardiographic data.This prohibited estimation of biological valve failure according to the guideline of the 2017 European Association of Percutaneous Cardiovascular Interventions/ European Society of Cardiology/European Association for Cardio-Thoracic Surgery consensus report because of the inability to identify structural valve deterioration not causing reintervention. 38Furthermore, proper evaluation of prosthesis-patient mismatch was not possible, information on functional status and quality of life was sparse, and duration of follow-up was limited.

CONCLUSIONS
In this study of more than 1000 unselected Freestyle full aortic root implantations in patients from 6 small-to medium-sized and comparable centers, survival was equal to that of previous reports of similar cohorts of aortic root procedures, and for elective indications also for that of the general population.
Considering the indications for Freestyle aortic root replacement and the known risk factors for death, the high early mortality rate in high-risk cases might be acceptable.Intermediate-term reintervention rates (valve and coronary) were low and primarily because of recurrent endocarditis.In conclusion, the present intermediate-term study provides further support for the use of the Freestyle full root bioprosthesis in the real-world setting of diverse, complex, and often high-risk aortic root replacement, and suggests that outcome is determined by patient and disease, rather than by prosthesis, characteristics.The Journal of Thoracic and Cardiovascular Surgery c December 2022 Adult: Aorta Dagneg ard et al

EARLY COMPLICATIONS
Early complications were defined as occurring within 30 days from surgery.We included all reoperations within 30 days for which the indication was drainage output, pericardial exudate, or tamponade symptoms.
Perioperative myocardial infarction was defined according to the Valve Academic Research Consortium-2 criteria E1 : New ischemic symptoms (eg, chest pain or shortness of breath), or New ischemic signs (eg, ventricular arrhythmias, new or worsening heart failure, new ST-segment changes, hemodynamic instability, new pathological Q-waves in at least 2 contiguous leads, imaging evidence of new loss of viable myocardium, or new wall motion abnormality), and Elevated cardiac biomarkers (preferable creatinine kinase myocardial band [CK-MB]) within 72 hours after the index procedure, consisting of at least 1 sample post procedure with a peak value exceeding 15 times the upper reference limit for troponin or 5 times for CK-MB.If cardiac biomarkers are increased at baseline (>99th percentile), a further increase in at least 50% post procedure is required and the peak value must exceed the previously stated limit.
If the diagnosis perioperative myocardial infarction was posed clinically, but information was not sufficient to assess the previously listed criteria, the event was included.
A postoperative cerebrovascular event was defined as Radiologically verified new cerebral lesion, or New neurological symptoms (not recurrence of previous symptoms) to ensure inclusion of all relevant events, also in the absence of the detailed information used in the Valve Academic Research Consortium-2 criteria.
Permanent pacemaker implantation: new implantation of a permanent pacemaker.

E-Reference
E1. Kappetein    In this study, an excessive mortality in patients who were operated for type A aortic dissection and endocarditis was shown, but a survival close to the general population in the primarily elective indications of aneurysm, small root, and "other" (consisting among other things of reoperative surgery and bailout surgery for failed aortic valve or root procedures).A subgroup analysis of men (A) and women (B) showed similar trends for both sexes.It appears as though female survival rates might be lower than that of the age-and sexstandardized general population, but because the confidence interval (CI) of all of the elective indications overlaps the curve of the general population for most of the period, this is not certain.A difference between sexes is shown in the group "others," however, taking into consideration the heterogenicity and low number of patients in this group, this alone cannot be interpreted as a true sex-mediated difference (C shows hazard ratios [HRs] obtained using Cox proportional hazards regression analysis per sex).The excessive mortality in the group "endocarditis" might be less pronounced in the female group but is nonetheless significantly worse than for aneurysm patients, in accordance with the male group (C).Further research is needed to further understand the influence of sex on aortic root replacement with the Freestyle bioprosthesis.Diss, Dissection; a-fib, atrial fibrillation; eGFR, estimated glomerular filtration rate. The Survival after aortic root replacement with the Freestyle bioprosthesis (Medtronic Inc, Minneapolis, Minn) versus the Danish general population.

FIGURE 2 .
FIGURE 2. Valve-related and coronary reinterventions occurring after unselected aortic root replacement with the Freestyle (Medtronic Inc, Minneapolis, Minn) bioprosthesis, according to indication of index Freestyle implantation.This observational study included 1008 patients from 6 North Atlantic centers.Freedom from reintervention was estimated by Aalen-Johansen analysis with death as competing risk (reported in FigureE4) stratified for indication for root replacement; curves are truncated at 10% remaining at risk in each indication group, vertical ticks represent censor points, and the 95% confidence interval (CI) is visualized as shading.A, Freedom from valve-related reinterventions including valve dysfunction, pseudoaneurysms, endocarditis, and thromboses requiring reintervention.B, Freedom from coronary reinterventions including percutaneous interventions and coronary artery bypass grafting.C, Crude and adjusted hazard ratios (HRs) obtained using Cox proportional hazards regression.Adjusted according to age at surgery, center for surgery, ischemic heart disease, atrial fibrillation (a-fib), estimated glomerular filtration rate (eGFR), valve size, and suture technique.

FIGURE 3 .
FIGURE 3. Study summary.This was a retrospective, observational North Atlantic multicenter study of survival and reinterventions after aortic root replacement with the stentless xenograft Freestyle (Medtronic Inc, Minneapolis, Minn) bioprosthesis.It included 1008 unselected patients from Copenhagen, Stockholm, Bergen, Montreal, Link€ oping, and Reykjavik.Results were stratified for the elective indications aneurysm, small root and other and emergent indications, complex endocarditis, and Stanford type A aortic dissection requiring root replacement.In this study survival differed significantly in between indications.Early mortality was high for acknowledged high-risk cases.For 90-day survivors, survival after elective indications resembled that of the sex-and age-standardized Danish general population, whereas those with type A dissection and endocarditis had continued excess mortality.Thus, prosthesis characteristics allow intermediate-term survival corresponding to the background population, and outcome after aortic root replacement with the Medtronic xenograft Freestyle bioprosthesis is determined by patient-rather than by prosthesis-characteristics.It is therefore useful, safe, and durable in the intermediate term in low-risk as well as complex and high-risk surgery.Type A Diss, Type A aortic dissection.1720TheJournal of Thoracic and Cardiovascular Surgery c December 2022

AFreestyle
FIGURE E2.All-cause survival for all and for 90-day survivors after aortic root replacement (ARR) with the Freestyle (Medtronic Inc, Minneapolis, Minn) bioprosthesis, compared with the Danish general population.Overall (A and B) and per indication: aneurysm (C and D), small root (E and F), other (G and H), endocarditis (I and J), and type A dissection (K and L).Shading indicates 95% confidence interval.Curves are truncated when 10% of the population remains at risk, vertical ticks represent censor points.Data for the general population were obtained from the publicly available Human Mortality Database (www.mortality.org), and standardized according to sex and age at year of surgery.

BFIGURE E4 .
FIGURE E4.Death before (A) valve-related and (B) coronary reintervention, as competing risk in the Aalen-Johnson model.Curves of "cause-2" (ie, death) from competing risk analyses of reinterventions after aortic root replacement with the Freestyle (Medtronic Inc, Minneapolis, Minn) bioprosthesis.Curves are truncated at 10% of the group population remaining at risk, censor times are marked with ticks and 95% confidence interval is indicated by shading.See Figure2for inverted cumulative incidence curves for (A) valve-related reinterventions and (B) coronary reinterventions.Aneurysm indicates aortic root aneurysm, small root indicates patients with aortic valve lesion at risk for patient-prosthesis mismatch, other represents primarily reoperative cases (with eg, aortic valve dysfunction, and bail-out for eg, aortic valve plasties), and endocarditis indicates complex endocarditis such as prosthetic endocarditis or root abscess.Diss, Dissection.
), most occurred among The Journal of Thoracic and Cardiovascular Surgery c December 2022

HR adjusted cox Small Root: Other: Endocarditis: Type A Dissection: Aneurysm: Small Root: Other: Endocarditis: Type A Dissection:
Landmark analysis with cutoff at 90 days postoperatively.C, Crude and adjusted hazard ratios (HRs) obtained using Cox proportional hazards regression for overall survival, truncated at 9.61 years where all indications have 10% of patients remaining at risk, at 90 days and at 0.25-9.61years, respectively.Diss, Dissection; a-fib, atrial fibrillation; eGFR, estimated glomerular filtration rate.
A C B FIGURE 1.Survival after unselected aortic root replacement with the Freestyle (Medtronic Inc, Minneapolis, Minn) bioprosthesis, according to indication.These 1008 patients underwent aortic root replacement with a stentless xenograft at 6 North Atlantic centers.Survival status follow-up was 100% complete.Survival was stratified for indication for root replacement; curves are truncated at 10% remaining at risk in each indication group, vertical ticks represent censor points and the 95% confidence interval (CI) is visualized as shading.The stapled line represents survival for the age-, sex-, and birth datestandardized Danish general population.A, Kaplan-Meier survival curves.B,

TABLE 3 .
Kaplan-Meier analysis of all-cause death after unselected aortic root replacement with the Freestyle bioprosthesis CI, Confidence interval; NA, nonapplicable.*Number at risk <10% and therefore not reported.The Journal of Thoracic and Cardiovascular Surgery c December 2022 (range, 2.2%-48.1%;only 1 patient had a EuroSCORE II<7%); hence our 30-day mortality rate is not completely unexpected but an over-representation of coronary reimplantation complications as a contributing factor cannot be out-ruled.Intermediate-term survival of patients with elective indications (aneurysm, small root, and other) was comparable or slightly lower than previously reported (Tables

TABLE 4 .
Early complications* after unselected aortic root replacement with the Freestyle bioprosthesis Intensive care unit; IQR, interquartile range.*Within 30 days from surgery.yVARC-2-criteria.zDefined as all new-onset neurological symptoms, including TIA.

TABLE 5 .
Excerpts from publications on ARR with specified indications and procedures: survival and reinterventions

TABLE 5 .
ContinuedThe table contains a selection of publications regarding ao root replacement, or alternative procedures to treat similar pathology.The table illustrates the differences in the baseline characteristics, including proportion of surgical priority and redo surgery, as well as the related results."Small root" indicates 21-mm annulus or prosthetic size.
"Repeat ARR" indicates second-time aortic root surgery.ARR, Aortic root replacement; SD, standard deviation; IQR, interquartile range; NYHA, New York Heart Association; CABG, coronary artery bypass grafting; redo, reoperative cardiac surgery (ie, redo sternotomy and dissection from adherences); VSRR, valve-sparing root replacement; El, elective; U, urgent; Hypoth, hypothermic; A-diss, type A aortic dissection; NA, Nonapplicable; AS, aortic stenosis; AR, aortic regurgitation; bio, bioprosthetic; mech, mechanical prosthesis; IE, infective endocarditis; AVR, aortic valve replacement; STS, The Society of Thoracic Surgeons; EScII, European System for Cardiac Operative Risk Evaluation II; Em, emergent; ca, circa; LOC, linearized occurrence rates; S, salvage; Ao, aortic; PVE, prosthetic valve endocarditis; CI, 95% confidence interval; IRAD, The International Registry of Acute Aortic Dissection; CRR, conservative root management; Prox, proximal (reoperations).*The number of patients on which the following numbers are based might be a subgroup of ARR in a study containing also other procedures.If the included number (and associated results) includes any of these other procedures it is reflected in the proportion of ARR, presented in the following column.yPercentage of the patients represented under "N" and the reported results -if the cited study contained subgroups that were not relevant and results were reported separately, only ARR cases are included in the table.zComposite end point with survival.x For hospital survivors.The Journal of Thoracic and Cardiovascular Surgery c December 2022

probability Time All aortic root replacement vs. Danish general population
AP, Head SJ, Genereux P, Piazza N, van Mieghem NM, Blackstone EH, et al.Updated standardized endpoint definitions for transcatheter aortic valve implantation: the Valve Academic Research Consortium-2 consensus document.J Thorac Cardiovasc Surg.2013;145:6-23.Distribution of indications for aortic root replacement with the Freestyle (Medtronic Inc, Minneapolis, Minn) bioprosthesis.All indications were represented at all institutions, but the distribution varied substantially.In the total study population the proportions were: aneurysm (aortic root aneurysm): 39.8% (n ¼ 410) , small root (patients with an aortic valve lesion at risk of prosthesis-patient mismatch): 8.3% (n ¼ 86), other (technically challenging cases such as redo aortic valve or root, or bailout surgery for aortic root remodeling or reimplantation): 13.8% (n ¼ 142), complex endocarditis (primarily prosthetic valve endocarditis or aortic root abscess) 26.7% (n ¼ 275), and Stanford type A aortic dissection (type A Diss), requiring root replace- The Journal of Thoracic and Cardiovascular Surgery c Volume 164, Number ment (eg, because of dissection involving the coronary or reaching the aortic valve): 11.4% (n ¼ 117).LNKPG, Link€ oping University Hospital; HUS, Haukeland University Hospital; IS, Landsp ıtali University Hospital of Iceland; MHI, Montreal Heart Institute; KS, Karolinska University Hospital; RH, Rigshospitalet, University Hospital of Copenhagen.1724.e2The Journal of Thoracic and Cardiovascular Surgery c December 2022 1724.e4The Journal of Thoracic and Cardiovascular Surgery c December 2022 Survival according to sex.In this study of 1008 patients who had undergone aortic root replacement with the Freestyle (Medtronic Inc, Minneapolis, Minn) bioprosthesis we investigated intermediate-term survival.

before valve-related (I) and coronary reintervention (II) respectively -as competing risk in Aalen-Johnson model. Absolute risk Years after surgery
Journal of Thoracic and Cardiovascular Surgery c Volume 164, Number 6 1724.e5

TABLE E1 .
Preoperative patient characteristics according to indication, including missing dataData are presented as % (n) except where otherwise noted.IQR, Interquartile range; BMI, body mass index; eGFR, estimated glomerular filtration rate; CABG, coronary artery bypass graft.
The Journal of Thoracic and Cardiovascular Surgery c Volume 164, Number Data are presented as % (n) except where otherwise noted.IQR, Interquartile range; BMI, body mass index; eGFR, estimated glomerular filtration rate; CABG, coronary artery bypass graft.1724.e8The Journal of Thoracic and Cardiovascular Surgery c December 2022

TABLE E4 .
Surgical characteristics according to indication, including missing data