Sex Differences in Clinical Outcomes and Loss of Life Expectancy After Aortic Valve Surgery for Infective Endocarditis: A SWEDEHEART Study Authors: Lisa Bearpark, Michael Dismorr, Anders Franco‐Cereceda, Ulrik Sartipy, Natalie Glaser Published: 2026 ABSTRACT BackgroundFemale patients with infective endocarditis (IE) are referred to surgery less often than male patients, and research on sex differences after IE surgery is inconclusive. Understanding the impact of sex can help clinicians optimize patient care. Therefore, we investigated sex‐based differences in loss of life expectancy and net survival after aortic valve surgery for IE.MethodsWe conducted a population‐based, observational cohort study, including all patients undergoing aortic valve surgery for IE in Sweden from 1997 to 2022. Data came from national registers, including SWEDEHEART (Swedish Web‐System for Enhancement and Development of Evidence‐Based Care in Heart‐Disease Evaluated According to Recommended Therapies). Loss of life expectancy and net survival were calculated based on population data from the Human Mortality Database. Flexible parametric models were used to assess net survival.ResultsA total of 2585 patients underwent aortic valve surgery for IE; 19% were female. Women were older than men (median age 66 versus 63) and had more concomitant valve surgery and comorbidities, as well as lower socioeconomic status. During a mean follow‐up of 7 years, 245 female patients (49%) versus 857 male patients (41%) died. Loss of life expectancy was accentuated in younger patients, ranging from 15.5 years (95% CI, 9.0–21.9) in a 50‐year‐old female to 2.1 years (95% CI, 0.95–3.2) in an 80‐year‐old male. After regression standardization, women had higher net survival compared with men.ConclusionsAfter aortic valve surgery for IE, net survival was higher among female patients after accounting for baseline differences, and loss of life expectancy was considerable for both female and male patients, particularly in younger patients. Focus on early diagnosis, as well as optimized pre‐ and postoperative care, is essential in patients with IE. Nonstandard Abbreviations and Acronyms infective endocarditis Swedish Web‐System for Enhancement and Development of Evidence‐Based Care in Heart Disease Evaluated According to Recommended Therapies Clinical PerspectiveWhat Is New? In this nationwide cohort of >2500 patients undergoing aortic valve surgery for infective endocarditis, female patients had higher net survival in relation to the Swedish population compared with male patients, after accounting for baseline differences.Loss of life expectancy was up to 16 years and was particularly accentuated among younger patients. What Are the Clinical Implications? Management of endocarditis should be sex independent and patients with a surgical indication should be referred to surgery irrespective of sex. Clinical Perspective What Is New? In this nationwide cohort of >2500 patients undergoing aortic valve surgery for infective endocarditis, female patients had higher net survival in relation to the Swedish population compared with male patients, after accounting for baseline differences.Loss of life expectancy was up to 16 years and was particularly accentuated among younger patients. In this nationwide cohort of >2500 patients undergoing aortic valve surgery for infective endocarditis, female patients had higher net survival in relation to the Swedish population compared with male patients, after accounting for baseline differences. In this nationwide cohort of >2500 patients undergoing aortic valve surgery for infective endocarditis, female patients had higher net survival in relation to the Swedish population compared with male patients, after accounting for baseline differences. Loss of life expectancy was up to 16 years and was particularly accentuated among younger patients. Loss of life expectancy was up to 16 years and was particularly accentuated among younger patients. What Are the Clinical Implications? Management of endocarditis should be sex independent and patients with a surgical indication should be referred to surgery irrespective of sex. Management of endocarditis should be sex independent and patients with a surgical indication should be referred to surgery irrespective of sex. Management of endocarditis should be sex independent and patients with a surgical indication should be referred to surgery irrespective of sex. Infective endocarditis (IE) is a complicated and serious disease with high mortality, and the incidence of IE is increasing. 1 , 2 , 3 IE is less commonly diagnosed in women, who comprise 30% to 40% of all cases 1 , 4 , 5 and only 20% to 25% of patients with IE who undergo surgery. 6 , 7 , 8 Female patients tend to present at an older age and exhibit a higher rate of mitral valve involvement, Staphylococcus aureus infection, and neurological symptoms. 8 , 9 , 10 Prior studies of sex differences in outcomes in patients with IE are inconclusive, with many reporting higher complication rates and worse outcomes in female patients. 9 , 11 , 12 , 13 Research has also found that female patients are less likely to be referred for valve surgery than their male counterparts, even after adjusting for baseline factors. 4 , 9 , 11 , 14 , 15 , 16 Patients with IE with a surgical indication who do not undergo surgery have worse outcomes. 16 , 17 Sex differences in surgically treated patients have been studied in single‐center settings, are based on small cohorts, or have limited follow‐up and report conflicting outcomes. 6 , 8 , 18 , 19 To our knowledge, none have analyzed net survival, which takes the longer life expectancy in women into consideration, despite known differences in expected survival between the sexes. 20 In this study, we examine sex‐related differences in net survival, all‐cause mortality, loss in life expectancy, heart failure hospitalization, and recurrent IE after aortic valve surgery for IE. Performed in a nationwide setting, our study aims to enhance the understanding of sex‐based differences in IE outcomes and inform presurgical risk‐stratification for patients undergoing aortic valve surgery for IE. METHODS Data Availability Due to Swedish law, we are not allowed to share the study database for this study. However, the data are available upon application from the different governing bodies named in the data sources section. Study Design We conducted a nationwide, observational cohort study with the approval of the Swedish Ethical Review Authority (original dnr: 2019‐04131; amendment dnr: 2020‐04967). The requirement for informed consent was waived. The Strengthening the Reporting of Observational Studies in Epidemiology guidelines were followed for the reporting of the study. 21 Setting About 600 to 800 patients are hospitalized annually in Sweden due to IE. According to the Swedish Register for Infective Endocarditis, which covers about 80% of hospitalized endocarditis cases in Sweden, female patients comprised 27% of all patients with IE in 2023. The overall surgical treatment rate has increased over time and was ∼40% in 2023 (an increase from 25% in 2008). 22 Data Sources We used the Swedish Cardiac Surgery register, a part of the SWEDEHEART (Swedish Web‐System for Enhancement and Development of Evidence‐Based Care in Heart‐Disease Evaluated According to Recommended Therapies) register, to identify the study cohort. 23 SWEDEHEART registers all patients undergoing cardiac surgery in Sweden. 24 Information is collected prospectively for >100 variables. Every Swedish citizen and individual living in Sweden has a unique personal number, which is used to index national health registers and enables cross‐linking of the registers. 25 For this study, SWEDEHEART was cross‐linked to the Swedish National Patient Register, 26 , 27 the Swedish Total Population Register, 27 and the longitudinal integrated database for health insurance and labor market studies register. 28 Baseline characteristics, including comorbidities, perioperative data, and socioeconomic factors, were sourced from these registers, which have been previously validated and described. 24 , 26 , 27 , 28 Follow‐up for all our outcomes was complete; an individual is lost‐to‐follow‐up only if they emigrate from Sweden, although most deaths of emigrated people are still reported with some delay. Study Population and Exposure We included all patients >18 years old who underwent aortic valve replacement with an IE indication in Sweden between January 1, 1997, and July 31, 2022. All of Sweden's 8 cardiac surgery centers were included. Exposure was defined as female sex as identified by the second‐to‐last number in the Swedish personal number, where an even number indicates female legal gender and an uneven number indicates male legal gender. 25 There were no exclusion criteria. All concomitant surgeries, including those of other valves, were included. Follow‐up for death ended on March 16, 2023, and follow‐up for heart failure hospitalization and recurrent IE ended on December 31, 2022. Definitions IE status at the time of surgery was identified using the European System for Cardiac Operative Risk Evaluation variable “active endocarditis” from the Swedish Cardiac Surgery register or the International Classification of Diseases, Tenth Revision (ICD‐10) codes I33.0, I33.9, I38.9, I39.8, and B37.6 from the National Patient Register, recorded at the time of hospital discharge. To avoid overestimating the number of IE cases, a hospital stay ≥14 days was required in conjunction with the IE diagnosis, except when death occurred <14 days of surgery. A list of definitions of comorbid conditions is presented in Table S1. Surgical Procedure Cardiac surgery for IE in Sweden is performed in accordance with European Society of Cardiology guidelines 29 with the aim of achieving radical debridement of infectious tissue. Type of surgery and choice of valve prosthesis are at the individual surgeon's discretion. Intraoperative transesophageal echocardiography is performed on all patients. Outcomes Vital status and date of death were sourced from the Total Population Register. 27 Expected survival was derived using Swedish life tables from the Human Mortality Database 30 and matched by age and year of surgery. Heart failure hospitalization and recurrent IE were sourced from the National Patient Register. 26 The validity of the National Patient Register has consistently been shown to be high, with a positive predictive value of 95% for the primary diagnosis of heart failure. 26 , 31 Early recurrent IE was defined as IE ≤1 year after index surgery, and late recurrent endocarditis as IE >1 year after index surgery. Table S2 lists the ICD codes used to identify the outcomes. For mortality, start of follow‐up was the date of surgery. To be certain to capture only postoperative heart failure and recurrent IE, heart failure hospitalization follow‐up started 90 days postoperatively, and recurrent IE follow‐up started 60 days postoperatively. Start of follow‐up for the respective outcomes was determined based on what made sense clinically and given the technical features of each register. For all outcome analyses, each participant contributed time until death, heart failure hospitalization or recurrent endocarditis, or end of follow‐up, whichever came first. Statistical Analysis Baseline characteristics are presented as frequencies and percentages for categorical values. Means±SD or median and interquartile range were used for continuous variables, depending on normalcy of distribution. We used the Kaplan–Meier method and the log‐rank test to compare crude survival. We used logistic regression to assess the relationship between sex and 30‐day mortality and reported age‐adjusted odds ratios (ORs) with 95% CIs. We used flexible parametric survival models (Royston–Parmar models) with regression standardization to fit smooth curves, allowing for the assessment of hazard over time and time‐varying effects. Visual inspection of Schoenfeld residuals was used to assess the proportional hazards assumption. Variance inflation factors were used to assess multicollinearity. The variables included in the final models (Data S1) were chosen based on prior research and clinical experience. The model was fitted using the stpm2 command in R and optimized based on both the Akaike information criterion and the Bayes information criterion. The final survival model had 4 degrees of freedom. The effects of age, emergency operation, biological prosthesis implantation, prior coronary artery bypass graft, and prior heart surgery were modeled as time‐dependent. Restricted cubic splines were used to model age, body mass index, and year of surgery. We plotted survival differences over time, presenting these visually. Loss of life expectancy was calculated using the observed survival among female and male patients in the study cohort, matched on age and year of surgery, to the expected survival in the Swedish population, obtained from life tables published in the Human Mortality Database (http://mortality.org). Net survival was defined as the difference in survival between the study population, stratified by sex and expected survival in the populace of Sweden (matched on age and year of surgery). In the adjusted model, we used regression standardization to account for intergroup differences between the patient groups (using the same stpm2 model as specified previously) and the calculated difference relative to the Swedish population as specified previously. The CuRe package in R was used for the analysis. 32 Flexible parametric models, using Markov multistate models as implemented in rstpm2 R‐package, that accounted for the competing risk of death were used to calculate heart failure hospitalization as well as recurrent IE. Multivariable Cox regression was used as sensitivity analysis for all outcomes, and we did 2 subgroup analyses, excluding first patients with prosthetic valve endocarditis and then patients who died within 30 days of surgery. A 2‐sided P value <0.05 was considered statistically significant. Data management and statistical analyses were performed using R statistical software version 4.3.1 (R Foundation for Statistical Computing, Vienna, Austria). Missing Data Missing variables were imputed using a classification and regression tree approach 33 as implemented by the Simputation package in R. Data were missing for body mass index (13%), prosthetic valve size (11%), left ventricular ejection fraction (15%), glomerular filtration rate (7%), and level of education (2%). RESULTS Baseline Characteristics Of 2585 patients who underwent surgery for aortic valve IE during the time of the study, 502 (19%) were female and 2083 (81%) were male. The number of surgeries performed increased progressively across the study period (Figure S1). Female patients were older than male patients (median age 66 versus 63 years). The median age at surgery increased in both sexes over the study period, from 64 to 68 years in women, and from 58 to 65 years in men, comparing surgeries performed 1997 to 2005 and 2015 to 2022, respectively. At baseline, more female patients had chronic obstructive pulmonary disease, reduced glomerular filtration rate, body mass index >30 kg/m2, and a lower household income. Perioperatively, more women had concomitant mitral (22% versus 17%) and tricuspid (7% versus 3%) valve surgery. Mechanical valve prosthesis implantation was more common in men (27% versus 34% in women versus men). Table 1 presents baseline characteristics and Table 2 presents peri‐ and postoperative factors, stratified by sex. Table 1Baseline Characteristics of 2585 Patients Who Underwent Aortic Valve Surgery for Infective Endocarditis in Sweden 1997 to 2022, Stratified by SexVariableOverallMale sexFemale sex P valueNo.25852083 (80.6)502 (19.4)Age, y, median (interquartile range)63.0 (51.0–71.0)63.0 (51.0–71.0)66.0 (54.2–73.0)<0.001Body mass index, kg/m2 <0.001<18.552 (2.3)29 (1.6)23 (5.4)18.5–24.9970 (43.3)788 (43.5)182 (42.5)25–29.9808 (36.1)687 (37.9)121 (28.3)>30410 (18.3)308 (17.0)102 (23.8)Non‐Nordic birth region212 (8.2)170 (8.2)42 (8.4)0.952Married1231 (47.7)1005 (48.3)226 (45.0)0.205Education0.919>12 y634 (24.9)514 (25.1)120 (24.3)10–12 y1125 (44.3)907 (44.3)218 (44.2)<10 y783 (30.8)628 (30.6)155 (31.4)Household income<0.001Q4 (max)645 (25.0)555 (26.7)90 (17.9)Q3646 (25.0)536 (25.8)110 (21.9)Q2646 (25.0)513 (24.7)133 (26.5)Q1 (min)646 (25.0)477 (22.9)169 (33.7)Prior endocarditis148 (5.7)126 (6.0)22 (4.4)0.182Prior cardiac surgery (any)674 (30.6)551 (30.9)123 (29.2)0.542Prosthetic valve endocarditis685 (26.5)565 (27.1)120 (23.9)0.259Prior surgical aortic valve replacement663549115Prior transcatheter aortic valve implantation16115Period of surgery, y0.2901997–2005597 (23.1)472 (22.7)125 (24.9)2006–2014880 (34.0)703 (33.7)177 (35.3)2015–20221108 (42.9)908 (43.6)200 (39.8)Hypertension1107 (42.8)884 (42.4)223 (44.4)0.450Prior atrial fibrillation677 (26.2)545 (26.2)132 (26.3)0.997Chronic obstructive pulmonary disease138 (5.3)96 (4.6)42 (8.4)0.001Hyperlipidemia426 (16.5)343 (16.5)83 (16.5)1.000Diabetes419 (16.2)336 (16.1)83 (16.5)0.879Prior myocardial infarction229 (8.9)185 (8.9)44 (8.8)1.000Prior percutaneous coronary intervention127 (4.9)107 (5.1)20 (4.0)0.338Prior heart failure746 (28.9)583 (28.0)163 (32.5)0.053Left ventricular ejection fraction0.217>50%1493 (67.9)1193 (67.0)300 (71.4)30%–50%622 (28.3)517 (29.0)105 (25.0)<30%85 (3.9)70 (3.9)15 (3.6)Prior pacemaker/implantable cardioverter‐defibrillator148 (5.7)117 (5.6)31 (6.2)0.707Liver disease95 (3.7)71 (3.4)24 (4.8)0.182History of cancer364 (14.1)295 (14.2)69 (13.7)0.865Preoperative bleeding event520 (20.1)423 (20.3)97 (19.3)0.666Prior stroke446 (17.3)347 (16.7)99 (19.7)0.118Peripheral vascular disease58 (2.2)49 (2.4)9 (1.8)0.554Chronic kidney disease148 (5.7)119 (5.7)29 (5.8)1.000Estimated glomerular filtration rate, mL/min per 1.73 m2 <0.001>601633 (67.6)1356 (69.8)277 (58.6)45–59310 (12.8)227 (11.7)83 (17.5)30–44223 (9.2)169 (8.7)54 (11.4)15–29116 (4.8)84 (4.3)32 (6.8)<15133 (5.5)106 (5.5)27 (5.7)Preoperative dialysis107 (4.4)86 (4.4)21 (4.4)1.000Alcohol dependence207 (8.0)177 (8.5)30 (6.0)0.076Opioid use56 (2.2)41 (2.0)15 (3.0)0.216Numbers are No. (%) unless otherwise noted. Baseline Characteristics of 2585 Patients Who Underwent Aortic Valve Surgery for Infective Endocarditis in Sweden 1997 to 2022, Stratified by Sex Numbers are No. (%) unless otherwise noted. Table 2Peri‐ and Postoperative Factors of 2585 Patients Who Underwent Aortic Valve Surgery for Infective Endocarditis in Sweden 1997 to 2022, Stratified by SexVariableOverallMale sexFemale sex P valueNo.25852083 (80.6)502 (19.4)Emergent operation347 (15.7)274 (15.4)73 (17.3)0.365Length of hospital stay, d, median (interquartile range)43.0 (33.0–59.0)43.0 (33.0–58.0)45.0 (34.0–65.0)0.016Mechanical valve835 (32.3)699 (33.6)136 (27.1)0.006Bioprosthetic valve1592 (61.6)1265 (60.7)327 (65.1)0.076Homograft65 (2.5)52 (2.5)13 (2.6)1.000Prosthetic valve size<0.00118–21519 (22.6)240 (12.9)279 (64.1)22–23841 (36.6)720 (38.6)121 (27.8)>23940 (40.9)905 (48.5)35 (8.0)Mitral valve surgery474 (18.3)362 (17.4)112 (22.3)0.012Tricuspid valve surgery86 (3.3)52 (2.5)34 (6.8)<0.001Coronary artery bypass grafting279 (10.8)225 (10.8)54 (10.8)1.000Ascending aorta replacement419 (16.2)337 (16.2)82 (16.3)0.986Root replacement473 (18.3)384 (18.4)89 (17.7)0.762Pacemaker/ICD implantation* 227 (8.8)182 (8.7)45 (9.0)0.942Acute kidney injury† 867 (42.6)701 (42.5)166 (43.3)0.804Numbers are No. (%) unless otherwise noted. ICD indicates implantable cardioverter defibrillator.*Defined as a new permanent pacemaker/ICD within 30 d of surgery. † Defined as >0.3 mg/dL (>26 μmol/L) increase in postoperative creatinine concentrations, postoperative creatinine >1.5× baseline, or new postoperative dialysis. Peri‐ and Postoperative Factors of 2585 Patients Who Underwent Aortic Valve Surgery for Infective Endocarditis in Sweden 1997 to 2022, Stratified by Sex Numbers are No. (%) unless otherwise noted. ICD indicates implantable cardioverter defibrillator. Defined as a new permanent pacemaker/ICD within 30 d of surgery. Defined as >0.3 mg/dL (>26 μmol/L) increase in postoperative creatinine concentrations, postoperative creatinine >1.5× baseline, or new postoperative dialysis. Survival During a mean follow‐up of 7.3 years (maximum 26.2 years), 1102 (43%) patients died, including 245 female patients (49%) and 857 male patients (41%). There was no statistically significant difference in 30‐day mortality between the sexes, which was 9% versus 7% among women and men (age‐adjusted OR, 1.14 [95% CI, 0.80–1.60]; P = 0.46). The crude all‐cause mortality (Table 3) at 1, 5, 10, and 15 years was 15%, 31%, 52%, and 63% among women, and 14%, 26%, 41%, and 53% among men. The Kaplan–Meier survival curves for female and male patients together with the age‐matched expected survival of women and men in the general population are presented in Figure 1. Although women had lower crude long‐term survival than men, this difference disappeared in the adjusted model. In fact, after regression standardization, the estimated survival difference at 15 years, was 4.5%, with higher survival among women (95% CI, 0.5%–8.5%). Crude net survival was lower among female patients (Figure 2), but, again, after accounting for differences in baseline characteristics between male and female patients, women had a slight statistically significant net survival advantage (Figure 3). The cumulative incidence of all‐cause mortality after regression standardization is presented in Table 4 and the crude and age‐adjusted incidence rates of all‐cause mortality are presented in Table S3. Table 3Crude Cumulative Incidence of All‐Cause Mortality, Heart Failure, and Recurrent Endocarditis % (95% CI)1 y5 y10 y15 yAll‐cause mortalityFemale sex14.8 (11.7 to 17.9)30.5 (26.2 to 34.7)51.5 (46.3 to 56.7)63.2 (57.4 to 68.9)Male sex13.7 (12.2 to 15.2)25.8 (23.8 to 27.8)40.8 (38.3 to 43.3)53.4 (50.5 to 56.4)Difference−1.1 (−4.5 to 2.4)−4.7 (−9.3 to 0.1)−10.7 (−16.3 to −4.7)−9.7 (−15.8 to −2.9)Heart failure hospitalizationFemale sex2.7 (1.2 to 4.1)9.3 (6.6 to 12.0)17.3 (13.4 to 21.2)20.0 (15.7 to 24.3)Male sex2.6 (1.9 to 3.3)6.4 (5.3 to 7.6)11.4 (9.8 to 13.0)16.8 (14.6 to 19.0)Difference−0.1 (−2.0 to 1.7)−3.9 (−7.5 to −0.2)−9.1 (−14.8 to −3.1)−5.1 (−12.1 to 2.5)Recurrent endocarditisFemale3.7 (2.0 to 5.4)8.8 (6.2 to 11.4)11.1 (8.1 to 14.0)13.4 (9.9 to 17.0)Male3.4 (2.6 to 4.2)8.4 (7.1 to 9.7)10.7 (9.2 to 12.2)13.4 (11.5 to 15.2)Difference−0.4 (−2.5 to 1.7)−0.7 (−4.1 to 2.8)−0.7 (−4.9 to 3.6)−1.0 (−7.2 to 5.5)All outcomes except all‐cause mortality used Aalen–Johansen estimator to account for the competing risk of death. Crude Cumulative Incidence of All‐Cause Mortality, Heart Failure, and Recurrent Endocarditis % (95% CI) All outcomes except all‐cause mortality used Aalen–Johansen estimator to account for the competing risk of death. Survival after aortic valve surgery for infective endocarditis in Sweden, 1997 to 2022, stratified by sex. Curves indicate Kaplan–Meier estimated survival for female and male patients with 95% CIs and expected survival for women and men in the general Swedish population, matched on age and year of surgery. Crude net survival among female and male patients. A, Net survival relative to the Swedish population (100%). Patients are matched to survival in the general population based on sex, age, and year of surgery, and presented with 95% CIs. B, Net survival difference over time between female and male patients. Presented with 95% CIs. Net survival after regression standardization among female and male patients. A, Net survival, after regression standardization, relative to the Swedish population (100%). Patients are matched to survival in the general population based on sex, age, and year of surgery, and presented with 95% CIs. B, Net survival difference over time between female and male patients after regression standardization. Presented with 95% CIs. Table 4Cumulative Incidence After Regression Standardization* of All‐Cause Mortality, Heart Failure Hospitalization, and Recurrent Endocarditis Stratified by Sex % (95% CI)1 y5 y10 y15 yAll‐cause mortalityMale sex14.5 (13.2 to 15.8)27.2 (25.6 to 28.9)44.3 (42.1 to 46.5)58.5 (56.0 to 61.1)Female sex12.5 (10.8 to 14.2)23.9 (21.2 to 26.6)39.9 (36.2 to 43.6)54.0 (49.9 to 58.2)Difference2.0 (0.3 to 3.7)3.3 (0.4 to 6.2)4.4 (0.5 to 8.3)4.5 (0.5 to 8.5)Heart failure hospitalization† Male sex0.7 (0.3 to 1.9)2.2 (0.9 to 5.6)4.6 (1.8 to 11.1)6.9 (2.8 to 16.5)Female sex0.8 (0.3 to 2.2)2.5 (1.0 to 6.3)5.2 (2.1 to 12.7)7.9 (3.2 to 18.9)Difference−0.1 (−0.3 to 0.1)−0.3 (−1.0 to 0.4)−0.6 (−2.0 to 0.8)−1.0 (−3.1 to 1.1)Recurrent endocarditis† Male sex2.5 (1.0 to 6.0)7.6 (3.3 to 17.2)10.3 (4.5 to 22.6)12.9 (5.7 to 27.8)Female sex2.7 (1.1 to 6.4)8.1 (3.5 to 18.2)10.9 (4.8 to 24.1)13.8 (6.0 to 29.7)Difference−0.2 (−0.9 to 0.6)−0.5 (−2.8 to 1.8)−0.7 (−3.7 to 2.4)−0.9 (−4.7 to 2.9)*The variables included in the models for the respective outcomes are detailed in Data S1. † Markov multistate models were used to account for the competing risk of death. Cumulative Incidence After Regression Standardization* of All‐Cause Mortality, Heart Failure Hospitalization, and Recurrent Endocarditis Stratified by Sex % (95% CI) The variables included in the models for the respective outcomes are detailed in Data S1. Markov multistate models were used to account for the competing risk of death. Loss of Life Expectancy Age‐ and sex‐matched loss of life expectancy was accentuated in younger patients (Figure 4). A 50‐year‐old female, undergoing surgery in 2000, had a loss of life expectancy of 15.5 years (95% CI, 9.05–21.9) and, if undergoing surgery the year 2020, a loss of life expectancy of 14.8 years (95% CI, 6.2–23.4), whereas a 50‐year‐old male, undergoing surgery in 2000, had a loss of life expectancy of 11.9 years (95% CI, 6.6–17.1) and, if undergoing surgery in 2020, a loss of life expectancy of 11.4 years (95% CI, 4.6–18.1). An 80‐year‐old female patient undergoing surgery in 2020 had a loss of life expectancy of 3.1 years (95% CI, 1.5–4.8), and an 80‐year‐old male patient undergoing surgery in 2020 had a loss of life expectancy of 2.1 years (95% CI, 0.95–3.2). Loss of life expectancy, in years, of female and male patients undergoing aortic valve surgery for Endocarditis Relative to the Swedish Population. Data are matched and stratified by sex, age groups, and year of surgery and presented with 95% CIs for female and male patients respectively. Heart Failure Hospitalization During a mean follow‐up of 6.8 years (maximum 25.9 years), 321 patients (12%) were hospitalized for heart failure, of whom 75 were female (15%) and 246 were male (12%). The crude cumulative incidence of heart failure at 1, 5, 10, and 15 years was 3%, 9%, 17%, and 20% among women and 3%, 6%, 11%, and 17% among men, respectively (Table 3). Women had significantly higher crude cumulative incidence of heart failure at 5 and 10 years (Table 3). After regression standardization, however, there was no statistically significant difference in heart failure hospitalization; the estimated difference at 15 years was −1.0% (95% CI, −3.1 to 1.1). The cumulative incidence of heart failure hospitalization after regression standardization is presented in Table 4 and Figure S2. The crude and age‐adjusted incidence rates of heart failure hospitalization are presented in Table S3. Recurrent IE During a mean follow‐up of 6.7 years (maximum 25.9 years), 269 patients (10%) had recurrent IE, of whom 54 were female (11%) and 215 were male (10%). Of those, 32% had early IE. Among all the patients with recurrent IE, 14 women (26%) and 72 men (33%) were reoperated. The cumulative incidence of recurrent IE at 1, 5, 10, and 15 years was 4%, 9%, 11%, and 13% among women and 3%, 8%, 11%, and 13% among men, respectively (Table 3). After regression standardization, the estimated difference in cumulative incidence of recurrent IE at 15 years was −0.9% (95% CI, −4.7 to 2.9). The age‐adjusted incidence rate of early recurrent endocarditis was 4 per 100 person‐years for both female and male patients, and the age‐adjusted incidence of late recurrent endocarditis was 1.3 and 1.2 per 100 person‐years for female and male patients, respectively (Table S3). The cumulative incidence of recurrent IE after regression standardization is presented in Table 4 and Figure S3. The crude and age‐adjusted incidence rates of recurrent IE are presented in Table S3. Sensitivity and Subgroup Analysis Cox regression gave similar results as the main analyses, as did our subgroup analyses, where we excluded patients who died within 30 days of surgery and patients with prosthetic valve IE. DISCUSSION In this nationwide study of patients undergoing aortic valve surgery for IE, with long and complete follow‐up, there were several important findings: (1) female patients tolerated surgery well, despite a worse preoperative risk profile; (2) the lower crude survival among female patients was explained by factors such as older age, more comorbidities, and multiple valve involvement, and adjusted net survival was higher among female than male patients; (3) female sex was not associated with increased rates of heart failure hospitalization or recurrent endocarditis after accounting for intergroup differences; and (4) both female and male patients operated for aortic valve IE had substantial loss of life expectancy, which was particularly accentuated among younger patients, with a loss of up to 14.2 years for a 50‐year old female patient operated in 2000. Few previous studies on outcomes after IE surgery have focused on aortic valve IE 6 , 8 , 18 , 19 and, to our knowledge, none have reported net survival or loss of life expectancy, taking the longer life expectancy of women into consideration. In a multicenter study, Ackermann et al. analyzed a cohort of 4917 patients (28% female) undergoing any valve surgery for IE; 3239 underwent aortic valve replacement (22% female). 19 There was no difference in 30‐day mortality between the sexes (14% versus 16%, among women versus men, P=0.066). In contrast to what we observed, however, Ackermann et al. found that female sex was an independent predictor of midterm mortality after multivariable adjustments (hazard ratio [HR], 1.20 [95% CI, 1.0–1.4], P=0.01). Possibly, this difference was attributable to the inclusion of IE surgery for any valve in the study by Ackermann et al. compared with our study, which focused on surgery for aortic valve engagement. Weber et al. conducted a single‐center study, analyzing 305 patients (25% female) undergoing surgery for IE, of whom 129 had aortic valve IE (18% women); results were reported as 30‐day and 1‐year survival (jointly for all valve surgeries). 8 More than twice as many women as men died within 30 days of surgery (18% versus 8%). Consistent with our study, the Weber et al. unadjusted analysis indicated lower long‐term survival for female patients, which disappeared after adjustment for baseline factors. Interestingly, unlike in our study, female patients were not significantly older than male patients. They also reported a higher incidence of acute kidney injury (37% versus 27%, P=0.008) and new pacemaker implantation among women compared with men (11% versus 3%, P=0.049), which we did not observe in our population (Table 2). In our study, we also report loss of life expectancy and sex‐based differences in net survival in relation to the general population, taking the longer life expectancy in women into consideration. Recently, Reitan et al. reported on loss of life expectancy in patients with myocardial infarction. 34 They found that loss of life expectancy was particularly attenuated in younger patients and in women, ranging from 11 years in 50‐year‐old women with reduced ejection fraction to 5 months in an 80‐year‐old male without reduced ejection fraction. We found a similar pattern, even more pronounced, for patients undergoing surgery for aortic valve IE. Younger patients had a greater loss of life expectancy than older patients, ranging from 14.2 years in a 50‐year‐old female to 2.1 years in an 80‐year‐old male patient. Our net survival analysis showed a statistically significant survival difference between men and women, but the regression‐standardized analysis revealed that differences in comorbidities and other baseline factors fully accounted for these differences. In the context of the studies discussed as well as others, 7 , 8 , 13 , 19 , 35 our study adds new sex‐specific information relevant to patients undergoing aortic valve surgery for IE and important information on loss of life expectancy for these patients. Our study also validates that female patients with IE have a different profile than male patients in terms of valve engagement, comorbidities, and socioeconomic factors. Despite potentially later referral as indicated in previous research, 4 , 9 , 14 , 15 the low postoperative mortality rate in our study reassuringly suggests that surgery was well tolerated also among female patients with elevated preoperative risk. The observation that female patients with IE have worse socioeconomic status corroborates the findings by Stahl et al. in Denmark 36 and is important, as socioeconomic factors are well known to affect cardiovascular health and health‐seeking behavior. 37 , 38 Clinicians must be aware of these differences, and more research on sex‐based differences at the time of diagnosis is warranted, especially as multiple studies have found delayed surgical referral in female IE patients. 4 , 9 , 14 , 15 In this study, women and men had similar long‐term survival, when accounting for preoperative differences, but both sexes suffered considerable loss of life expectancy, underscoring the importance of early diagnosis and urgent referral to surgery when indicated. Strengths and Limitations The strengths of this study are the comprehensive cohort, sourced from high‐quality Swedish registers, and the rigorous data on >30 baseline and perioperative factors. Although the register‐based study design allows a uniquely large and population‐based study cohort including detailed patient data with long and complete follow‐up, the registers used lack specific information on pathogens and echocardiographic data such as vegetations and paravalvular complications. We did, however, use root replacement as a proxy for more invasive disease, and there was no difference in proportion of root replacements between the sexes. We also did not know if patients with concomitant mitral and tricuspid valve procedures were infected in each valve or not, and we did not have information on patients who were not operated, which may have introduced surgical selection bias. Other important outcome measures such as quality of life and functional capacity were not available. Although we adjusted for a comprehensive set of baseline variables, the observational nature of the study means that some degree of residual confounding is likely to persist. CONCLUSIONS This nationwide study of patients undergoing aortic valve surgery for IE confirms that IE remains a serious disease with considerable loss of life expectancy for both female and male patients. However, although crude survival was lower among women, female patients had slightly higher net survival than male patients after accounting for baseline differences. There were no differences in heart failure hospitalization or recurrent IE. Female patients undergoing surgery for aortic valve IE were older and at higher surgical risk compared with male patients. Our findings underscore the importance of early diagnosis and timely management of infective endocarditis, both pre‐ and postoperatively. Sources of Funding This work received support from the Swedish Heart‐Lung Foundation (grants 20241040, 20220364 and 20190570 to Dr Glaser, grant 2024133 to Dr Dismorr, and grant 20220151 to Dr Sartipy), Region Stockholm clinical postdoctoral appointment (grant FoUI‐955489 to Dr Glaser), the Magnus Bergvall Foundation (grants 2023–576 and 2022–177 to Dr Glaser), the Mats Kleberg Foundation (grant 2022–119 to Dr Glaser), Karolinska Institutet Foundations and Funds (grants 2024‐02855 and 2022‐01575 to Dr Glaser, grant 2024‐02499 to Dr Sartipy, and grant 2024‐03099 to Dr Dismorr), and a donation from Mr Fredrik Lundberg and the Schörling Foundation (to Dr Franco‐Cereceda). The supporting bodies had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the article; and decision to submit the article for publication. Disclosures None. Supporting information Data S1. STROBESEX Tables S1–S3 Figures S1–S3 Supplemental Methods