Aortic Valve Replacement With Mechanical Valves vs Perimount Bioprostheses in 50- to 69-Year-Old Patients

Background Evidence is mixed regarding the most appropriate type of valve prosthesis for surgical aortic valve replacement (AVR) in patients 50 to 69 years. American and European guidelines differ. Objectives The purpose of this study was to determine the long-term all-cause mortality and complication rates after AVR in patients aged 50 to 69 years according to implantation of a Perimount bioprosthetic valve or a mechanical valve. Methods In this nationwide observational cohort study, all patients aged 50 to 69 years who underwent primary surgical AVR in Sweden 2003 to 2018 using a Perimount bioprosthesis or mechanical valve were identified from the SWEDEHEART register. Primary outcome; all-cause mortality, secondary outcomes; major bleeding, aortic valve reintervention, heart failure hospitalization, and stroke. National health-data registers were used to ascertain outcomes. Regression standardization addressed confounding. Results A total of 6,907 patients aged 50 to 69 years were included (Perimount group, n = 3,831 and mechanical valve group, n = 3,076) and 74% were men. The use of bioprostheses increased during the study period. At 15 years of follow-up, the estimated cumulative incidence of all-cause mortality was 37% (95% CI: 35%-40%) vs 45% (95% CI: 42%-48%) in the mechanical and Perimount groups, respectively (survival difference −7.9% [95% CI: −11% to −4.6%]). Patients with mechanical valves had a lower risk of aortic valve reintervention but a higher risk for bleeding. Survival difference at 15 years in ages 50 to 59 years was −15% (95% CI: −8.4% to −21%). Conclusions In patients aged 50 to 69 years who underwent surgical AVR, survival was better in those who received mechanical compared to Perimount bioprosthetic valves. While valve choice should be guided by individual patient factors and patient preference rather than by chronological age, the substantial survival advantage observed in patients with mechanical valves in ages 50 to 59 years must be recognized.

U ntreated severe aortic valve steno- sis has an annual mortality rate of up to 25% and a mean survival of 2 to 3 years. 1,2Aortic valve replacement (AVR) is the only definitive treatment and can be performed by open surgery or transcatheter aortic valve replacement (TAVR).Although AVR is life-saving, the loss of life expectancy following AVR is approximately 3 to 4 years in patients aged 50 to 69 years when compared with that in the general population. 3It has been debated whether a mechanical or biological aortic valve prosthesis should be used in patients aged 50 to 69 years, 4 and the results of previous studies have been contradictory, [5][6][7] leading to different recommendations in the American and European guidelines. 8,9Comparative long-term performance studies of aortic bioprostheses have shown wide variation among the various bioprosthetic valve models. 10,11In a previous study using data from the SWEDEHEART (Swedish Web-system for Enhancement and Development of Evidence-based care in Heart disease Evaluated According to Recommended Therapies) register, 12,13 we observed that survival was better in patients aged 50 to 69 years who underwent surgical AVR using a mechanical valve than in those who received a bioprosthetic valve. 6In a subsequent SWEDEHEART study, we found that long-term performance and survival was better when a Carpentier-Edwards PERIMOUNT valve (Edwards Lifesciences Corp), was placed than when other bioprosthetic valves were used. 11In view of our results, 6,11 we thought it worthwhile to investigate if the best bioprosthesis could outperform a mechanical valve in patients aged 50 to 69 years who require AVR.The Swedish Cardiac Surgery Register records all patients who have undergone cardiac surgery in Sweden since 1992 and contains preoperative, perioperative, and postoperative data, including survival status.The Swedish Cardiac Surgery Register has been shown to have high reliability and validity. 13The Swedish National Patient Register was used to obtain baseline characteristics, including comorbidities and bleeding, hospitalization for heart failure, aortic valve reintervention, and stroke outcomes.The heart failure and stroke diagnoses in the National Patient Register have also been shown to have high reliability and high validity. 16,17Socioeconomic background characteristics were obtained from the longitudinal integrated database for health insurance and the labor market studies, maintained by Statistics Sweden. 18The Swedish personal identity number made it possible to cross-link data at the individual level. 19These national registers have been described elsewhere. 6POSURE AND OUTCOMES.The patients were categorized as having either a mechanical valve or a Perimount bioprosthesis.The Perimount bioprostheses included the 2900, Magna 3000, and Magna Ease 3300 models.The valve models and their frequency of use are listed in Supplemental Table 1.
The primary outcome was all-cause mortality, as obtained from the Swedish Population Register. 20 continuous variables as the mean AE SD.Time to event was defined as the number of days from the date of surgery until the date of the event or the end of follow-up.The end of follow-up for death was March 17, 2020 and that for the secondary outcomes was December 31, 2018.The Kaplan-Meier method was used to calculate the crude survival.The Aalen-Johansen estimator was used to estimate the crude cumulative incidence of bleeding, heart failure hospitalization, reintervention, and stroke while accounting for the competing risk of death.A Poisson model was used to obtain age-and sex-adjusted incidence rates.To account for differences at baseline, the standardized cumulative survival and differences in survival were estimated using flexible parametric regression standardization.The resulting survival curve estimates the population outcome if the entire population would have received each respective type of valve.This method adjusts for the population distribution of covariates. 21,22Model selection was performed using clinical subject matter knowledge and a backward selection strategy informed by the Akaikes information criterion.Details of the final models are available in the Supplemental Appendix.Flexible hazard-based regression standardization was used to estimate the cumulative incidence and differences in bleeding, reintervention, heart failure hospitalization, and stroke rates as described by Kipourou et al. 23 The resulting cumulative incidence curve estimates the population outcome, had the entire population received one or the other type of valve.This method adjusts for the population distribution of covariates while accounting for the competing risk of death.The Classification and Regression Tree estimation and imputation approach was used to handle missing data. 24As a sensitivity analysis, the main analyses were repeated using an inverse probability of treatment weighting (IPTW) approach.Stabilized weights were obtained using generalized boosted regression models. 25We also performed subgroup analyses according to whether patients were aged 50 to 59 years or 60 to 69 years and in the subset of patients who underwent isolated AVR.All statistical analyses and data management were performed using the R programming language, version 4.2.0 (R Foundation for Statistical Computing), and included the use of the "mexhaz," "rstpm2," and "WeightIt" packages. 21,26,27 1.

In
CLINICAL OUTCOMES.The crude and age-and sexadjusted incidence rates for all outcomes are shown in Supplemental Table 3.The regression standardized cumulative incidence of all outcomes at 15 years is shown in Table 2 and at 5 and 10 years in Supplemental Tables 4 and 5.   of follow-up, the estimated cumulative incidence of stroke was similar in both groups (14%, 95% CI: 12%-16%).The between-group difference in cumulative incidence of stroke was 0.1% (95% CI: À2.3% to 2.7%).
The regression standardized cumulative incidence of stroke is shown in Figure 2.  6 and 7.The regression standardized survival and survival differences in age groups 50 to 59 years and 60 to 69 years is shown in Table 2 and Figure 3.The regression standardized complication rates are shown in Table 2 and Supplemental Figures 5 and 6.We also conducted analyses using IPTW to control for confounding factors; the results were very similar to those of the main analyses that used regression standardization.All analyses were repeated in the subset of patients who underwent isolated AVR, and the findings were consistent with the findings in the total study population.

DISCUSSION
In this study, we found that survival was better after Our findings should be interpreted in light of previous studies that investigated clinical outcomes in patients aged 50 to 69 years who underwent surgical AVR with a mechanical valve or a bioprosthesis because the evidence is mixed.Some studies found that long-term survival was significantly better in patients aged 50 to 70 years who received a mechanical valve rather than a bioprosthetic valve, 6,28,29 whereas others found no significant survival difference. 5,7Chiang et al 5 performed an observational study in patients aged 50 to 69 years, who underwent primary isolated AVR in New York State between 1997  Goldstone et al 7 investigated 9,942 patients who underwent primary, isolated AVR in the state of California between 1996 and 2013 and were followed up until December 31, 2013.After IPTW, the 15-year mortality was significantly higher in patients aged 45 to 54 years who received a bioprosthesis than in those who received a mechanical valve but there was no survival difference in those aged 55 to 64 years.
Our previous nationwide study using data from the SWEDEHEART registry included patients aged 50 to 69 years who underwent primary isolated AVR between 1997 and 2013 and were followed up until Comparison between regression-standardized cumulative incidence of complications in patients who received a mechanical valve and those who received a Perimount bioprosthesis.The shaded areas show the 95% CI.The curves represent the estimated cumulative incidence of the respective event, if the entire population had received a mechanical valve or a Perimount valve, respectively.Bleeding was significantly greater and aortic valve reintervention was significantly less common in the mechanical valve group than in the Perimount group.There was no significant difference in the cumulative incidence of heart failure or stroke between the groups (curves superimposed).
Lu et al Mechanical or Bioprosthetic AVR in Ages 50 to 69 Years J U N E 2 0 2 3 : 1 0 0 3 5 9 March 24, 2014.In 1099 propensity score-matched patient pairs, we found that survival was better with a mechanical valve than with a bioprosthetic valve.
However, subgroup analyses suggested that the survival benefit was restricted to patients aged 50 to 59 years. 6A study from Cleveland Clinic included all patients who underwent isolated AVR from 1990 to January 2020. 30Perioperative outcomes were compared in 527 propensity score matched pairs.
Early postoperative clinical outcomes were similar between the groups.Adjusted long-term survival in the total study population (5,506 patients with bioprostheses and 637 patients with mechanical valves) was not statistically different between the groups.
However, the study did not specifically address outcomes in ages 50 to 69 years, and there was a clear institutional preference for bioprosthetic rather than mechanical valves.[7] Our results support the current European Society of Cardiology/European Association for Cardio-Thoracic Surgery recommendation that a mechanical prosthesis should be considered for the aortic position in patients aged younger than 60 years. 9The American College of Cardiology/American Heart Association guideline for the management of valvular Our study also has some limitations.First, we were not able to ascertain secondary outcomes (aortic valve reintervention, major bleeding events, heart failure hospitalization, or stroke) that occurred outside of Sweden.However, in view of the universal tax-financed health care coverage in Sweden, these were likely to have been infrequent, and it is reasonable to assume that these events occurred nondifferentially between the groups.
Therefore, this limitation may lead to underestimation of event rates but is unlikely to bias the results.Deaths outside of Sweden were captured by the Population Register, and follow-up regarding the main outcome was thus complete. 20Second, although our characterization of the study population was granular, we lacked information regarding potentially important patient features (eg, frailty) that would be associated with both exposure (valve choice) and outcome (all-cause mortality).Thus, residual confounding could have affected our findings.Third, three Perimount models (2900, Magna [3000], and Magna Ease [3300]) were categorized into 1 group.It is possible that performance may differ between these three models. 32Fourth, the Swedish national health-data registers do not contain echocardiographic data.Therefore, we could not ascertain structural valve deterioration according to current definitions 33,34 and relied on surrogate measures (aortic valve reintervention and heart failure hospitalization).

CONCLUSIONS
We found that in patients aged 50 to 69 years who underwent surgical AVR, survival was better in those who received a mechanical valve than in those who METHODSSTUDY DESIGN.The protocol for this observational, nationwide, population-based cohort study was approved by the Swedish Ethical Review Authority.The requirement for informed consent was waived because the data were deidentified.Study reporting followed the STrengthening the Reporting of OBservational studies in Epidemiology and REporting of studies Conducted using Observational Routinely collected health Data guidelines. 14,15STUDY POPULATION AND DATA SOURCES.All patients aged 50 to 69 years in Sweden who underwent surgical AVR with a Perimount bioprosthesis or mechanical valve with or without concomitant coronary artery bypass grafting or ascending aortic surgery between January 1, 2003, and December 31, 2018 were included in the study.Patients were excluded if they met any of the following criteria: age younger than 50 years or older than 69 years, previous cardiac surgery, previous TAVR, concomitant surgery on another valve, use of deep hypothermia and circulatory arrest, undetermined type of prosthesis, or use of a bioprosthetic valve other than a Perimount.The study population was identified from the Swedish Cardiac Surgery register, which is a part of the SWE-DEHEART registry.

3
Secondary outcomes were the cumulative incidence of bleeding, aortic valve reintervention (defined as a subsequent surgical AVR or TAVR), heart failure hospitalization, and stroke.The International Classification of Diseases-10 codes used to ascertain each outcome were obtained from the National Patient Register and are presented in Supplemental Table 2. STATISTICAL ANALYSIS.Categorical variables are presented as the frequency (percentage) and A B B R E V I A T I O N S A N D A C R O N Y M S AVR = aortic valve replacement CI = confidence interval IPTW = inverse probability of treatment weighting TAVR = transcatheter aortic valve replacement Lu et al J A C C : A D V A N C E S , V O L . 2 , N O . 4 , 2 0 2 Mechanical or Bioprosthetic AVR in Ages 50 to 69 Years total, 6,907 patients aged 50 to 69 years who underwent surgical AVR in Sweden during the study period and fulfilled the inclusion criteria were identified (Perimount group, n ¼ 3,831; mechanical valve group, n ¼ 3,076).The numbers of mechanical and Perimount AVRs carried out per year in Sweden during the study period are shown in Supplemental

STROKE.
During a mean follow-up of 6.1 years (maximum 16.0), 502 patients (7.3%) experienced a stroke event.The crude cumulative incidence of stroke is shown in Supplemental Figure 4.At 15 years SENSITIVITY AND SUBGROUP ANALYSES.When the clinical outcomes were reassessed after stratification according to age group 50 to 59 years or 60 to 69 years, the results were consistent with the findings in the total study population (Central Illustration).The baseline characteristics in the two age groups are shown in Supplemental Tables surgical AVR in patients aged 50 to 69 years who received a mechanical valve than in those who received a Perimount bioprosthetic valve.Patients with a mechanical valve had a lower risk of needing aortic valve reintervention but had a higher risk of bleeding complications.No significant betweengroup difference was found in the risk of hospitalization for heart failure or of stroke.These findings were consistent in patients aged 50 to 59 years and those aged 60 to 69 years.However, in patients aged 60 to 69 years, the absolute survival difference at 15 years was attenuated, while the increased risk of major bleeding remained approximately the same as that in younger patients.While valve choice should be guided by individual patient factors and patient preference rather than by chronological age, the substantial survival advantage observed at the population level in patients with a mechanical valve in the age group 50 to 59 years must be recognized.

FIGURE 1
FIGURE 1 Survival and Differences in Survival

J
A C C : A D V A N C E S , V O L . 2 , N O . 4 Mechanical or Bioprosthetic AVR in Ages 50 to 69 Years and 2004 and analyzed 1001 propensity score matched patient pairs.Follow-up ended on November 30, 2013, and the authors concluded that there was no significant difference in 15-year survival.

FIGURE 2
FIGURE 2 Cumulative Incidence of Complications CENTRAL ILLUSTRATION Substantial Survival Difference in Ages 50 to 59 Years Lu R, et al.JACC Adv.2023;2(4):100359.Can the best bioprosthesis outperform a mechanical valve for surgical AVR in patients aged 50 to 69 years?In 6,907 patients aged 50 to 69 years who underwent surgical AVR 2003 to 2018, survival was better after mechanical vs Perimount bioprosthetic AVR.While valve choice should be guided by individual patient factors and patient preference rather than by chronological age, the substantial survival advantage observed in patients with mechanical valves in ages 50 to 59 years must be recognized.AVR ¼ aortic valve reintervention.J A C C : A D V A N C E S , V O L . 2 , N O . 4 Mechanical or Bioprosthetic AVR in Ages 50 to 69 Years referral single center experience, and it is unclear to what extent the results are generalizable to other institutions.COMPLICATIONS AFTER AVR.The findings of previous studies have been largely concordant regarding complications following AVR and concluded that patients who received a mechanical valve have a higher cumulative incidence of bleeding but a lower risk of reoperation. 31Notably, Goldstone et al concluded that bioprostheses were associated with fewer strokes in patients aged 45 to 54 years, while the studies by Chiang et al and Glaser et al found no difference in

FIGURE 3 3
FIGURE 3 Survival and Differences in Survival in Age Groups 50 to 59 and 60 to 69 Years received a Perimount bioprosthesis.Patients with a mechanical valve had less risk of aortic valve reintervention but a higher risk of bleeding complications.The survival advantage associated with a mechanical valve was substantial in patients aged 50 to 59 years, suggesting that it may be prudent to adjust the current American College of Cardiology/ American Heart Association valvular heart disease guideline to recommend a mechanical valve for patients up to 60 years of age.

TABLE 1
Baseline Characteristics in Patients Aged 50 to 69 Who Underwent Surgical Aortic Valve Replacement With a Perimount or Mechanical Valve Mechanical or Bioprosthetic AVR in Ages 50 to 69 Years Figure 1.The number of implants per year according to subtype of Perimount bioprosthesis are shown in Supplemental Figure 2. The mean age was 61.9 years (standard deviation 5.2) and 74% of the study population were men.There were small but potentially important baseline differences between the groups; eg, the mean age was 64 years in the Perimount group and