Abstract
BACKGROUND:
Mitral valve (MV) infective endocarditis (IE) often necessitates surgical intervention for severe mitral regurgitation, yet the longitudinal impact of MV surgery within a repair-first strategy remains unclear.
METHODS:
An institutional multidisciplinary endocarditis database incorporating Society of Thoracic Surgeons data identified patients undergoing MV surgery for acute IE between July 2016 and June 2024. Patients were managed with a repair-first operative strategy, with MV replacement reserved for cases in which repair was not feasible. Primary outcomes were survival and the composite of reinfection, reoperation, or death. Secondary measures included perioperative outcomes and the individual components of the composite.
RESULTS:
198 total patients underwent first-time MV intervention for infective endocarditis [Repair 102 (51.5%)] or replacement [96 (48.5%)]. The repair group was younger (45 vs 51 years, p=0.03) and had fewer prior non-mitral cardiac surgeries (18.6% vs 34.4%, p=0.01). The repair group had less prolonged ventilation (14.7% vs 32.3%, p=0.003) and shorter intensive care unit stay (73 vs 121 hours, p<0.0001). Mitral valve repair was associated with superior risk-adjusted survival (HR 0.49, p=0.01) and freedom from the composite outcome of reoperation, reinfection, or death (HR 0.27, p<0.0001).
CONCLUSIONS:
Complex MV repair (defined as leaflet modification and/or patch reconstruction) in acute IE is safe and durable. Within a repair-first strategy, successful repair was associated with favorable perioperative recovery and improved mid-term outcomes. These findings support a repair-first approach when anatomically and technically feasible.
Acute infective endocarditis (IE) involving the mitral valve (MV) is highly morbid and often requires surgical intervention. Surgical options include repair or replacement, but practice patterns vary widely.1 Single center series suggest repair is feasible in 30% to 80% of cases depending on valve destruction and surgical expertise.2 Registry analyses suggest a survival advantage with repair; the Danish registry demonstrated replacement was associated with a significantly higher 10-year risk of recurrent IE, while the Dutch registry identified replacement as an independent predictor of worse longitudinal survival after surgery for active IE.3,4 In the United States, administrative datasets, such as the National Inpatient Sample, report that only 25% of MV operations for IE are repairs, despite evidence that repair is associated with lower in-hospital morbidity and mortality compared to replacement.5
Despite these insights, practice patterns continue to vary significantly, and a lack of evidence-based guidance remains. Most available longitudinal outcomes data come from European populations, limiting generalizability to U.S. cohorts where the epidemiology of IE differs.3,4 In particular, the ongoing opioid epidemic has fueled a surge in intravenous drug–associated (IVDA) IE creating a younger, higher-risk surgical population.6 While registry studies have demonstrated feasibility and benefits of repair, existing U.S. data are limited to administrative datasets that report low repair rates and capture only perioperative outcomes.5 To date, there are no longitudinal comparative studies from U.S. cohorts, particularly in referral centers managing high volumes of IE in patients who inject drugs.
The present study was designed to evaluate perioperative and mid-term outcomes of MV surgery for acute IE within a repair-first strategy at a high-volume comprehensive valve center. We specifically examined outcomes among patients successfully undergoing repair versus those requiring replacement when durable repair was not feasible. We hypothesized that patients amenable to successful repair, even when complex, would demonstrate favorable perioperative recovery and improved mid-term outcomes.
PATIENTS AND METHODS
Patient Data
Consecutive adult patients (age >18) who underwent cardiac surgery for acute IE by experienced mitral surgeons in a comprehensive valve center between July 2016 and June 2024 were included (n=741). All standard Society of Thoracic Surgeons (STS) data variables were integrated into an institutional multidisciplinary database. Additional data abstraction was performed to obtain supplementary endocarditis information including valve specifics and repair type. Patients with prior mitral valve intervention were excluded. Complex mitral valve repair was defined as repair requiring leaflet modification and/or patch reconstruction. West Virginia University Health Sciences Institutional Review Board approval was obtained with waiver of consent (Protocol #1709755537, Approved 6/23/22).
Clinical Care
All patients were managed by a multidisciplinary endocarditis care team including a cardiac surgeon, cardiologist, infectious disease specialist, palliative care expert, addiction medicine specialist, medical ethicist, social worker, and hospital internal medicine physician or critical care physician. Management was standardized for all patients and those who met criteria for surgical intervention for acute IE were scheduled for the next available operating room time with the intent of a repair-first approach. Operations were performed by six surgeons, all of whom have incorporated mitral valve repair into their practice.
Consistent with our repair-first approach, mitral valve replacement was reserved for valves deemed not amenable to durable repair based on intraoperative assessment of the extent of infection and leaflet destruction. Repair was pursued whenever a durable reconstruction could be achieved, including cases requiring complex techniques such as leaflet augmentation, chordal replacement, or annular reconstruction. Intraoperative conversion from attempted repair to replacement was not specifically captured as a distinct variable in the present dataset but was uncommon. Robotic approaches were utilized in selected patients based on anatomic suitability and procedural considerations. While robotic cases in this cohort were more commonly isolated mitral valve procedures, robotic techniques are applied across multiple valve pathologies and reflect patient selection rather than a distinct treatment strategy. Use of robotic approaches was consistent across the study period.
Postoperative management was standardized, and the multidisciplinary endocarditis care team continued to provide recommendations on antibiotic management and disposition.
Follow Up
Follow-up was obtained through longitudinal chart review within a regional integrated health system encompassing 38 cardiology sites with centrally maintained echocardiographic data. Patients underwent routine clinical follow-up with serial echocardiography, typically on an annual basis. Additional outcomes were captured through review of all available inpatient and outpatient encounters within the health system.
Statistical Analysis
Patients were stratified by MV repair or replacement when durable repair was not feasible. Continuous variables were reported as mean ± standard deviation, or median [Q1, Q3] for non-parametric data, and categorical data was summarized as number (%). The Mann-Whitney U Test compared continuous variables and Chi-Square Test for categorical variables. Primary outcomes included survival, and the composite of reoperation, reinfection, or death. Secondary analyses included perioperative outcomes and individual components of the composite outcome. Risk-adjustment was performed using multivariable regression models. Covariates were chosen a priori based on clinical relevance and established associations with outcomes in infective endocarditis and mitral valve surgery. Variables included patient demographics, major medical comorbidities, endocarditis characteristics, prior cardiac surgery, and operative complexity. Operative variables (e.g., cardiopulmonary bypass and cross-clamp times) were included as surrogates of procedural complexity rather than independent predictors of risk. A full list of covariates included in the adjusted models is provided in Supplemental Table 1. Analyses were conducted with Cox proportional hazards regression and Fine Gray model with competing risk of death. Models were evaluated with likelihood-ratio tests to prevent overfitting. All statistical analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC).
RESULTS
Patient Characteristics
A total of 198 consecutive patients undergoing MV surgery for acute IE were analyzed [102 (51.5%) repair, 96 (48.5%) replacement] (Table 1). Patients undergoing repair were younger (45 vs 51 years, p=0.03), more frequently had substance use disorder (62.8% vs 45.8%, p=0.02), and were less likely to have prior cardiac surgery (18.6% vs 34.4%, p=0.01). STS predicted risk of mortality was 4.2% in the replacement group and 1.2% in the repair group (p<0.0001), consistent with greater baseline operative risk in the replacement cohort. Rates of diabetes, renal dysfunction, and cerebrovascular disease were similar between groups, and other comorbidities including hypertension, heart failure, and lung disease were comparable.
Table 1.
Baseline Characteristics and Operative Details
| Total (n=198) |
MV Repair (n=102) |
MV Replacement (n=96) |
P-value | |
|---|---|---|---|---|
| PREOPERATIVE | ||||
| Age | 47.5 (33, 62) | 45 (31, 59) | 51 (35.5, 64.5) | .0292 |
| Male | 112 (56.57%) | 65 (63.73%) | 47 (48.96%) | .0362 |
| BMI | 26.98 (22.23, 32.73) | 26.98 (22.13, 32.07) | 27.07 (22.69, 33.55) | .5680 |
| Substance Use Disorder | 108 (54.55%) | 64 (62.75%) | 44 (45.83%) | .0169 |
| Ejection Fraction | 60 (55, 63) | 60 (55, 63) | 60 (55, 63) | .5043 |
| Creatinine | 0.87 (0.69, 1.16) | 0.84 (0.69, 1.12) | 0.94 (0.695, 1.25) | .1364 |
| Heart Failure | 75 (37.88%) | 42 (41.18%) | 33 (34.38%) | .3241 |
| Dialysis | 12 (6.06%) | 7 (6.86%) | 5 (5.21%) | .6258 |
| Diabetes | 53 (26.77%) | 22 (21.57%) | 31 (32.29%) | .0885 |
| Lung Disease | 32 (16.41%) | 13 (13.00%) | 19 (20.00%) | .2648 |
| Active Endocarditis | 171 (86.36%) | 85 (83.33%) | 86 (89.58%) | .2003 |
| Prior Sternotomy | 52 (26.26%) | 19 (18.63%) | 33 (34.38%) | .0119 |
| Prior Valve Replacement | 32 (16.16%) | 4 (3.92%) | 28 (29.17%) | <.0001 |
| Mild MR | 29 (14.72%) | 16 (15.84%) | 13 (13.54%) | .2463 |
| Moderate MR | 33 (16.75%) | 16 (15.84%) | 17 (17.71%) | .2463 |
| Severe MR | 127 (64.47%) | 68 (67.33%) | 59 (61.46%) | .2463 |
| STS Predicted Mortality | 2.6 (1.1, 5.9) | 1.2 (0.7, 2.2) | 4.2 (2.8, 9.3) | <.0001 |
| OPERATIVE | ||||
| AV Repair | 5 (2.53%) | 1 (0.98%) | 4 (4.17%) | .1532 |
| AV Replace | 36 (18.18%) | 21 (20.59%) | 15 (15.63%) | .3655 |
| TV Repair | 31 (15.66%) | 19 (18.63%) | 12 (12.50%) | .2357 |
| TV Replace | 12 (6.06%) | 6 (5.88%) | 6 (6.25%) | .9137 |
| Robotic Surgery | 27 (13.64%) | 23 (22.55%) | 4 (4.17%) | .0002 |
| Perfusion Time | 146 (104, 199) | 152 (117, 196) | 133 (88, 210.5) | .0560 |
| Clamp Time | 112 (74, 147) | 118 (86, 153) | 103 (64, 140) | .0509 |
AV: Aortic Valve, MR: Mitral Regurgitation, TV: Tricuspid Valve
Values are presented as n (%) or median (IQR)
Single-valve MV surgery was performed in 114 patients (57.6%), while 84 (42.4%) underwent multi-valve procedures involving the aortic or tricuspid valves. Multi-valve intervention rates were similar between repair and replacement cohorts (45.1% vs 36.5%). All patients demonstrated significant leaflet destruction with involvement of ≥1 scallop; vegetation size was similar between groups (16mm vs 18mm, p=0.16). Complex reconstruction was common, with autologous untreated pericardial patch augmentation utilized in 55% of repairs. Among patients requiring patch reconstruction, patch position was anterior in 23 (36.5%), posterior in 19 (30.2%), and bridging in 21 (33.3%). All patches utilized autologous untreated pericardium. Post-repair hemodynamics were favorable, with all patients demonstrating ≤1+ residual MR and transmitral gradient ≤4 mmHg. Mean follow-up was 2.8 years (range 0.7–7.8).
Perioperative Outcomes
Compared with replacement, patients undergoing MV repair had lower rates of prolonged ventilation (14.7% vs 32.3%, p=0.003), shorter intensive care unit (ICU) length of stay (72.9 vs 120.6 hours, p<0.0001), and fewer STS major morbidity events (20.6% vs 39.6%, p=0.004) (Table 2).
Table 2.
Perioperative Outcomes
| Total (n=198) |
MV Repair (n=102) |
MV Replacement (n=96) |
P-value | |
|---|---|---|---|---|
| Transfusion | 140 (70.71%) | 61 (59.80%) | 79 (82.29%) | .0005 |
| Atrial Fibrillation | 36 (18.18%) | 17 (16.67%) | 19 (19.79%) | .5688 |
| Reoperation | 23 (11.62%) | 8 (7.84%) | 15 (15.63%) | .0877 |
| Renal Failure | 17 (8.59%) | 5 (4.90%) | 12 (12.50%) | .0565 |
| Prolonged Ventilation | 46 (23.23%) | 15 (14.71%) | 31 (32.29%) | .0034 |
| Stroke | 4 (2.02%) | 0 (0.00%) | 4 (4.17%) | .0373 |
| ICU stay (hours) | 85.45 (48.9, 162.7) | 72.85 (40.9, 129.6) | 120.6 (66.6, 231.75) | <.0001 |
| Length of Stay | 19 (8, 37) | 16.5 (7, 36) | 20 (10, 40.5) | .2631 |
| STS Major Morbidity | 59 (29.80%) | 21 (20.59%) | 38 (39.58%) | .0035 |
| Operative Mortality | 16 (8.08%) | 6 (5.88%) | 10 (10.42%) | .2420 |
| STS Morbidity/Mortality | 62 (31.31%) | 22 (21.57%) | 40 (41.67%) | .0023 |
| 30-Day Readmission | 33 (16.67%) | 10 (9.80%) | 23 (23.96%) | .0076 |
ICU: Intensive Care Unit, STS: Society of Thoracic Surgeons,
Values are presented as n (%) or median (IQR)
Longitudinal Outcomes
Longitudinally, patients undergoing repair had lower incidence of reinfection (12.8% vs 30.2%, p=0.003), fewer reoperations (7.8% vs 17.7%, p=0.04), and lower all-cause mortality (23.5% vs 44.8%, p=0.011) (Table 3). Kaplan–Meier analysis showed improved survival (log-rank p = 0.01), as well as greater freedom from reinfection (log-rank p=0.002) (Figures 1 and 2).
Table 3.
Mid-Term Outcomes
| Total (n=198) |
MV Repair (n=102) |
MV Replacement (n=96) |
P-value | |
|---|---|---|---|---|
| Moderate or Severe MR | 11 (5.56%) | 4 (3.92%) | 7 (7.29%) | .3008 |
| Paravalvular Leak | 5 (2.53%) | NA | 5 (5.21%) | NA |
| Structural Valve Deterioration | 12 (6.06%) | NA | 12 (12.50%) | NA |
| Recurrent Drug Use | 21 (12.57%) | 10 (11.49%) | 11 (13.75%) | .6605 |
| Reinfection | 42 (21.21%) | 13 (12.75%) | 29 (30.21%) | .0027 |
| Reoperation | 25 (12.63%) | 8 (7.84%) | 17 (17.71%) | .0367 |
| Mortality | 67 (33.84%) | 24 (23.53%) | 43 (44.79%) | .0016 |
| Follow Up Time (Years) | 2.41 (0.54, 4.72) | 2.53 (0.83, 4.66) | 2.19 (0.49, 4.82) | .7976 |
MR: Mitral Regurgitation
Values are presented as n (%) or median (IQR)
Figure 1.
Kaplan-Meier Survival Curve
Figure 2.
Freedom from Reinfection
After risk adjustment, MV repair was associated with a lower rate of the composite endpoint of reoperation, reinfection, or death (HR 0.27, p<.001) and improved overall survival (HR 0.49, p=0.01) compared to MV replacement.
To evaluate whether concomitant procedures influenced these findings, an additional analysis was performed in patients undergoing isolated mitral valve operations. Within this subgroup, repair remained associated with a trend toward improved survival (log-rank p=0.09) and significantly greater freedom from reoperation (log-rank p=0.02) (Supplemental Figures 1 and 2).
COMMENT
The present study evaluated outcomes of mitral valve repair versus replacement for acute IE within a repair-first strategy, yielding three principal findings. First, complex repair was frequently achievable by experienced surgeons—even in the setting of substantial leaflet destruction—often requiring patch augmentation and achieving uniformly favorable immediate hemodynamics (≤1+ MR; low transmitral gradients). Second, repair was associated with a more favorable perioperative course, including less prolonged ventilation, shorter ICU stay, and fewer major morbidity events than replacement. Third, over longitudinal follow-up, patients undergoing repair had lower rates of reinfection, reoperation, and all-cause mortality, with greater freedom from the risk-adjusted composite endpoint of reinfection, reoperation, and death. Collectively, these data demonstrate that when anatomically feasible, mitral valve repair is safe and associated with favorable mid-term survival and durability in the setting of acute IE.
Successful reconstruction of severely damaged mitral valves relies on surgical experience and programmatic volume, which may translate into reproducible repair even in anatomically destructive or infected settings. National analyses from the Society of Thoracic Surgeons Adult Cardiac Surgery Database demonstrate a clear volume–outcome relationship for mitral valve surgery, with hospitals performing more than 75 mitral procedures annually and surgeons performing at least 35 cases per year achieving significantly lower 30-day mortality (0.7% vs 1.4%) and higher repair rates (84% vs 64%) compared with the lowest-volume quartiles.7 In high-volume repair programs, such as those performing more than 150 repairs per year, even lower-volume surgeons achieve excellent outcomes within the framework of an experienced institutional ecosystem, though conversion from planned repair to replacement remains more common among less experienced operators (9.0% vs 3.4%).8 These findings support the importance of concentrated experience and multidisciplinary collaboration in expanding the anatomic boundaries of repair. Within such comprehensive valve programs, complex reconstruction—including patch augmentation, chordal replacement, and annular reconstruction—can be performed safely and reproducibly. Consistent with these observations, our institution achieved high repair feasibility (over half of all mitral IE cases) and excellent durability, reinforcing the role of specialized experience in optimizing outcomes for this challenging pathology.
Young patients with IVDA endocarditis represent a particularly challenging subgroup, characterized by recurrent infection, continued substance use, and markedly worse longitudinal survival compared with non-IVDA populations.9,10 Prosthetic valve replacement in this cohort may not be a definitive solution as mechanical and bioprosthetic valves demonstrate similar risks of reinfection and structural failure, while mechanical prostheses introduce additional challenges related to anticoagulation management and adherence.11 Moreover, prosthetic durability is inherently limited, particularly in younger patients whose higher cardiac output and inflammatory burden may accelerate degeneration.12 Re-operative valve surgery in this setting carries higher operative risk and worse outcomes than first-time intervention.13,14 Accordingly, achieving a durable repair at the index operation is beneficial, and our findings suggest that when durable repair is achievable, mid-term durability, survival, and freedom from reinfection are favorable.
This study has several limitations. It reflects experience from a single, high-volume comprehensive valve center with a well-established mitral valve repair program, which may limit generalizability. Although multivariable adjustment was performed, residual confounding remains possible, particularly since valve selection depends on anatomic feasibility and surgeon judgment. Patients undergoing mitral valve replacement in this cohort represent those with more extensive valvular destruction and higher disease burden not fully captured in retrospective analyses, and thus comparisons between groups should be interpreted within this context. Cardiopulmonary bypass time and cross-clamp times were used as surrogates for case complexity, but can be affected by many factors. Additionally, follow-up was limited to a mean of 2.8 years, and long-term durability, especially among young patients, remains uncertain. Finally, the study population reflects the epidemiologic and socioeconomic characteristics of the Appalachia region of the United States, where the burden of intravenous drug use, limited access to longitudinal care, and social vulnerability may influence both outcomes and follow-up completeness.
In conclusion, mitral valve repair for acute infective endocarditis was frequently achievable within a repair-first strategy and was associated with favorable perioperative recovery, lower rates of reinfection and reoperation, and improved mid-term survival compared to patients who underwent replacement. These findings support a repair-first approach in experienced centers capable of complex reconstruction.
Supplementary Material
ABBREVIATIONS
- AV
Aortic Valve
- ICU
Intensive Care Unit
- IE
Infective Endocarditis
- IVDA
Intravenous Drug-Associated
- MR
Mitral Regurgitation
- MV
Mitral Valve
- STS
Society of Thoracic Surgeons
- TV
Tricuspid Valve
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Declaration of interests
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
REFERENCES
- 1.Evans CF, Gammie JS. Surgical management of mitral valve infective endocarditis. Semin Thorac Cardiovasc Surg. 2011;23(3):232–240. [DOI] [PubMed] [Google Scholar]
- 2.Scheggi V, Olivotto I, Del Pace S, et al. Feasibility and outcome of mitral valve repair in patients with infective endocarditis. Cardiothorac Surg. 2020;28:27. [Google Scholar]
- 3.Alhakak A, Butt JH, Havers-Borgersen E, et al. Mitral valve replacement or repair and long-term risk of infective endocarditis: a Danish nationwide study. Eur Heart J. Published online July 11, 2025. [DOI] [PubMed] [Google Scholar]
- 4.Tomšič A, de Weger A, van der Stoel M, Klautz RJM, Palmen M; Cardiothoracic Surgery Registration Committee of the Netherlands Heart Registration. A Nationwide Study on Mitral Valve Repair vs Replacement for Active Endocarditis. Ann Thorac Surg. 2024;117(1):120–126. [DOI] [PubMed] [Google Scholar]
- 5.Alkhouli M, Alqahtani F, Berzingi C, Cook CC. Contemporary trends and outcomes of mitral valve surgery for infective endocarditis. J Card Surg. 2019;34(7):583–590. [DOI] [PubMed] [Google Scholar]
- 6.Cook CC, Rankin JS, Roberts HG, et al. The opioid epidemic and intravenous drug-associated endocarditis: A path forward. J Thorac Cardiovasc Surg. 2020;159(4):1273–1278. [DOI] [PubMed] [Google Scholar]
- 7.Badhwar V, Vemulapalli S, Mack MA, et al. Volume-Outcome Association of Mitral Valve Surgery in the United States. JAMA Cardiol. 2020;5(10):1092–1101. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Newell P, Percy E, Hirji S, et al. Outcomes of Mitral Valve Repair Among High- and Low-Volume Surgeons Within a High-Volume Institution. Ann Thorac Surg. 2023;115(2):412–419. [DOI] [PubMed] [Google Scholar]
- 9.Straw S, Baig MW, Gillott R, et al. Long-term Outcomes Are Poor in Intravenous Drug Users Following Infective Endocarditis, Even After Surgery. Clin Infect Dis. 2020;71(3):564–571. [DOI] [PubMed] [Google Scholar]
- 10.Goodman-Meza D, Weiss RE, Gamboa S, et al. Long term surgical outcomes for infective endocarditis in people who inject drugs: a systematic review and meta-analysis. BMC Infect Dis. 2019;19(1):918. Published 2019 Nov 8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Anantha-Narayanan M, Reddy YNV, Sundaram V, et al. Endocarditis risk with bioprosthetic and mechanical valves: systematic review and meta-analysis. Heart. 2020;106(18):1413–1419. [DOI] [PubMed] [Google Scholar]
- 12.Xue Y, Kossar AP, Abramov A, et al. Age-related enhanced degeneration of bioprosthetic valves due to leaflet calcification, tissue crosslinking, and structural changes. Cardiovasc Res. 2023;119(1):302–315. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Rajjoub H, Wolfe S, Lagazzi LF, et al. Recurrent substance use and reoperative valve surgery for acute infective endocarditis. JTCVS Open. 2025;25:89–95. Published 2025 May 2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Mori M, Bin Mahmood SU, Schranz AJ, et al. Risk of reoperative valve surgery for endocarditis associated with drug use. J Thorac Cardiovasc Surg. 2020;159(4):1262–1268.e2. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.