Mid-term outcomes of a modified prosthetic aortic valve implantation technique for prosthetic valve endocarditis complicated by aortic annular abscess

Abstract

Background

Prosthetic valve endocarditis (PVE) complicated by complex paraannular aortic abscess remains a surgical challenge. We aimed to evaluate the mid-term outcomes of a standardized, patch-free, supra-annular aortic valve replacement strategy that avoids complete root replacement in these high-risk patients.

Methods

This retrospective, single-center study included 70 consecutive patients treated between 2015 and 2023 for complex aortic PVE using a consistent surgical protocol: radical debridement, supra-annular prosthetic valve implantation, inside the Valsalva sinuses, and external felt reinforcement—without annular or root patching. Follow-up included annual clinical and echocardiographic assessment (mean duration, 5.2 ± 1.3 years). The primary endpoint was freedom from the composite of reinfection or aortic reoperation. Secondary endpoints included survival, valve performance, and predictors of late adverse events.

Results

Hospital mortality was 5.7% (n = 4 of 70). At 5 years, overall survival was 86.0% (95% confidence interval [CI], 78.0%-93.9%), and freedom from reinfection or aortic reoperation was 89.2% (95% CI, 82.1%-96.2%). Outcomes were comparable between the original (n = 47) and expanded (n = 23) cohorts (log-rank P = .71). Two late reinfections (2.9%) were managed medically. One patient required reoperation for structural valve degeneration at 54 months. Mean transvalvular gradient remained stable at 5 years (10.4 ± 3.1 mm Hg at discharge vs 11.1 ± 3.6 mm Hg; P = .27) whereas mean left ventricular ejection fraction improved significantly (from 52 ± 10% to 58 ± 8%; P < .01). Ninety-two percent of survivors were in New York Heart Association class I-II. Independent predictors of late events included EuroSCORE II >12% and incomplete abscess excision.

Conclusions

Patch-free supra-annular valve replacement offers durable infection control and excellent hemodynamic outcomes, avoiding the need for complete root replacement in complex aortic PVE.

Graphical Abstract

For selected cases of prosthetic aortic valve endocarditis complicated by annular abscess, we propose a simplified surgical technique without the use of a pericardial patch, with prosthesis implantation in the supra-annular position. The 5-year follow-up results are very encouraging.

1A, Supra-annular anchoring mattress. 2A, Valve seating; 3A-5A, Circumferential reinforcement.

Central Message.

A patch-free, supra-annular valve replacement offers durable infection control and excellent mid-term outcomes in complex prosthetic valve endocarditis, avoiding the need for annular or root reconstruction.

Perspective.

Complex prosthetic valve endocarditis with para-annular abscess poses major surgical challenges. Traditional methods often require root replacement or annular patching. This study shows that a standardized technique—radical debridement, supra-annular valve inside sinuses, and external felt—yields excellent mid-term results, avoiding complex reconstruction and reducing the risk of reinfection.

Prosthetic valve endocarditis (PVE) represents one of the gravest challenges in contemporary cardiac surgery, accounting for 20% to 30% of all infective endocarditis (IE) cases and carrying in-hospital mortality rates frequently exceeding 25%. The presence of an infected prosthetic surface, often compounded by extensive para-annular abscess formation, hampers effective antibiotic penetration, and mandates radical surgical debridement for source control. Para-annular aortic abscesses complicate up to 40% of PVE episodes and are independently associated with higher rates of perioperative mortality and late recurrence.¹ Traditional surgical approaches—homograft root replacement, cryopreserved allografts, or Bentall–De Bono procedures—provide satisfactory infection eradication but are hindered by limited homograft availability, technical complexity in hostile redo fields, prolonged myocardial ischemia, and difficulties in future coronary access.²

In 2021, we described a modified prosthetic valve implantation technique designed to combine radical debridement with simplified reconstruction. The key innovation lies in supra-annular implantation of a prosthetic valve without the use of annular patches or complete root replacement, thereby reducing cross-clamp time and avoiding coronary button reimplantation when dense adhesions are present. Early results in 47 patients demonstrated encouraging freedom from reinfection and excellent hemodynamic performance at 2-year follow-up; however, midterm durability and generalizability remained uncertain.³

Since that initial report, several developments have sparked renewed interest in alternative strategies for aortic root reconstruction in complex PVE:

• •Large multicenter series have confirmed >90% 5-year survival with homograft root replacement, although at the expense of longer cardiopulmonary bypass (CPB) times and limited graft supply.⁴
• •Sutureless and rapid-deployment valves have simplified explant-and-replace paradigms, but concerns persist regarding sizing constraints and residual infective foci within the annulus.⁵
• •Hybrid endocarditis boards and 3-dimensional imaging now facilitate tailored surgical plans, underscoring the need for versatile techniques adaptable to varying degrees of tissue destruction.⁶

Against this backdrop, we aimed to expand our initial cohort by including 23 additional consecutive patients treated between 2020 and 2023, thereby providing a comprehensive assessment of 70 patients managed with the same modified technique strategy and followed for a mean of >5 years.

This study had 3 specific objectives: (1) to report updated perioperative outcomes, including mortality, neurologic events, and technical complications; (2) to analyze long-term freedom from the composite endpoint of reinfection or aortic reoperation; and (3) to identify independent predictors of late adverse events. We hypothesized that extending the follow-up to 5 years and nearly doubling the sample size would confirm the durability of infection control and valvular function observed in the initial series while also substantiating the procedure's reproducibility across evolving surgical teams and perioperative protocols.

Methods

Study Design and Setting

We conducted this retrospective, single-center cohort study at Anthea Hospital GVM Care & Research, Cardiac Surgery Department, a tertiary referral center for IE and complex aortic root surgery. The study was approved by Institutional Review Board (PVE MOD 2025 01, approved January 7, 2025) and was conducted in compliance with the Declaration of Helsinki. Written informed consent for surgical treatment, data collection, and long-term follow-up was obtained from each patient or their legal representative.

Patient Selection

All consecutive adult patients (age >18 years) who underwent to a new replacement for infected aortic prosthesis with the modified prosthetic implantation technique between January 2015 and December 2023 were screened (Figure E1). All patients provided consent for the disclosure of scientific data and publication at the time of surgical consent.

Figure E1.

Flowchart of patient selection.

Use of the technique was considered only when there was intraoperative confirmation of disconnection of the aortomitral junction (ie, disruption of the aortomitral curtain) with annular abscess because of the need for PVE reconstruction, and when a stable circumferential anchoring zone could be achieved after radical debridement. Absolute contraindications were (1) near-circumferential annular or multiple sinus root destruction preventing secure fixation; (2) left main coronary artery (LM) ostial involvement or anatomic configurations (eg, low-lying/short LM) in which the ostium would be jeopardized by the suture line and could not be safely mobilized or protected; (3) uncontrolled extension to the aortomitral curtain requiring alternative root/double-patch reconstruction; and (4) multiple sinus destruction or extensive aortocavitary fistulas. Relative contraindications included active fungal PVE/persistent bacteremia favoring homograft root replacement, porcelain aorta, and severely friable tissues unlikely to hold pledgeted sutures.

Preoperative Imaging and Coronary Risk Assessment

All patients underwent electrocardiographic-gated contrast computed tomography (CT) scan using a TAVI style protocol (LM/right coronary artery ostial height, sinus of Valsalva diameters, sinotubular junction, leaflet length, LM takeoff/angulation) and comprehensive TEE. Anatomies at risk for ostial encroachment (eg, low-lying or short LM, small sinuses, unfavorable sinotubular junction geometry) were considered contraindications to the present technique and managed with root replacement.

Surgical Protocol

The step-by-step modified root implantation technique (Figure 1) has been described in a previous article.³ Key technical principles include the following:

• •Leave the abscess cavity open to the aortic lumen for continuous postoperative drainage, enhancing antibiotic penetration.
• •Avoid additional prosthetic material inside the root; the only foreign bodies are the sewing ring and felt strip.
• •Maintain cross-clamp time <100 minutes by limiting complications.

Figure 1.

Surgical protocol, step-by-step. 1A, Supra-annular anchoring mattress sutures. 2A, Valve seating. 3A-5A, External circumferential reinforcement.

Postrepair TEE documented a competent valve with no or trivial paravalvular leak (PVL) prior to weaning from CPB.

The aortopulmonary window was addressed with limited anterior separation of the aorta from the pulmonary artery sufficient to expose the noncoronary sinus and aortomitral junction. Circumferential root mobilization was not performed, and the LM was not routinely dissected or looped. LM protection relied on subcoronary valve implantation (ie, sewing ring below the ostia) and intra-aortic direct inspection of the LM ostium before completion of the annular suture line; conversion to root replacement was undertaken if ostial clearance could not be guaranteed.

Circumferential reinforcing bands were not used. Reinforcement was limited to the noncoronary sinus suture line (ie, targeted suture line buttressing) and was not placed posteriorly behind the coronary ostia or across commissures. Interrupted pledgeted nonabsorbable sutures were oriented away from the LM ostium.

Data Collection

Demographic, clinical, microbiological, intraoperative, and outcome data were retrieved from our prospectively maintained endocarditis registry. Variables captured included age, sex, EuroSCORE II, causative organism, abscess size, presence of concomitant mitral involvement, CPB/cross-clamp times, need for coronary button reimplantation, and completeness of abscess excision (graded by the operating surgeon).

Follow-up Protocol

Patients were reviewed at 6 weeks, 6 months, and annually thereafter. Clinical evaluation included New York Heart Association (NYHA) classification, blood cultures if febrile episodes occurred, and transthoracic echocardiography assessing prosthetic gradient, PVL, and left ventricular ejection fraction (LVEF). Follow-up was 100% complete (mean, 5.2 ± 1.3 years; median, 5.0 years; interquartile range [IQR], 4.1-6.3 years).

Study Endpoints

The primary endpoint was freedom from the composite of definite PVE reinfection (according to Duke criteria plus positive cultures) or aortic reoperation for any cause at 5 years. Secondary endpoints included overall survival, cardiovascular mortality, NYHA functional status, major adverse cardiovascular events (eg, stroke, acute myocardial infarction, renal failure requiring dialysis), prosthetic valve hemodynamics (eg, mean gradient, prosthesis–patient mismatch), and the need for permanent pacemaker implantation.

Statistical Analysis

Continuous variables are expressed as mean ± standard deviation (SD) or median (IQR) and compared using the Student t test or Mann-Whitney U test, as appropriate. Categorical variables are presented as count (percentage) and compared using the χ² or Fisher exact test. Kaplan-Meier curves were constructed for the primary endpoint and overall survival; differences between the original (n = 47) and expansion (n = 23) cohorts were assessed with the log-rank test. Cox proportional hazards modelling was used to identify independent predictors of the composite endpoint, entering variables with P < .10 on univariable analysis. Proportionality assumptions were verified with Schoenfeld residuals. Statistical significance was defined as P < .05 (2-sided). Analyses were performed with SPSS v28 (IBM) and R 4.3.1 (survival package).

Results

Baseline Characteristics

Patients meeting any absolute contraindication (as defined in Methods) were treated with root replacement and were not included in the present cohort (Table 1). The cohort included 70 patients (mean age, 64.8 ± 11.2 years; 71% male) with complex PVE and para-annular abscess. The median EuroSCORE II value was 11.3 (IQR, 7.2-18.4), and blood cultures were positive in 58 patients (83%). Staphylococcus species predominated (n = 26; 37%), followed by Enterococcus (n = 11; 16%) and Candida (n = 3; 4%). PVL at admission was documented in 52 patients (74%), and 29 patients (41%) were in NYHA class III/IV.

Table 1.

Baseline characteristics

Variable	Total (N = 70)	Original (N = 47)	Expansion (N = 23)
Age, y, mean ± SD	64.8 ± 11.2	65.1 ± 11.4	64.2 ± 10.9
Male sex, n (%)	50 (71)	34 (72)	16 (70)
EuroSCORE II, median (IQR)	11.3 (7.2-18.4)	11.5 (7.3-18.9)	10.9 (7.0-17.8)
PALSUSE score, median (IQR)	7 (6-9)	7 (6-8)	8 (7-9)
Diabetes mellitus, n (%)	20 (29)	9 (19)	9 (39)
Abscess diameter, mm, mean ± SD	13.6 ± 4.2	13.4 ± 4.1	14.0 ± 4.4
Cross-clamp time, min, mean ± SD	95.63 ± 38.96	94.96 ± 41.02	97 ± 35.20
CPB time, min, mean ± SD	171.02 ± 73.26	170.03 ± 74.78	173.03 ± 71.65
Biological prosthesis, n (%)	38 (54.3)	25 (53.2)	13 (56.5)
Mechanical prosthesis, n (%)	32 (45.7)	22 (46.8)	10 (43.5)

SD, Standard deviation; IQR, interquartile range; CPB, cardiopulmonary bypass.

The expansion group (n = 23) was overall comparable to the original cohort (n = 47) across all key clinical and surgical parameters. Minor differences were observed, with patients in the expansion group being slightly younger (mean, 62 ± 12 years vs 66 ± 10 years; P = .046) and more frequently diabetic (39% vs 19%; P = .048); however, abscess dimensions, LVEF, valve type, and infecting organism were essentially identical. The median interval from index aortic valve replacement to PVE diagnosis was 24 months (IQR, 12-37 months). Surgery on the ascending aorta, limited to the segment distal to the sinotubular junction and sparing the aortic root, was required in 8 of 70 patients (11%), owing mainly to poor tissue quality (n = 6) or aortic dilation >45 mm (n = 2).

Operative Details and Early Outcomes

Mean CPB and cross-clamp times were 118 ± 34 minutes and 94 ± 28 minutes, respectively (Table 2). No patient required complete aortic root replacement or coronary button reimplantation; however, in 8 patients (11%), replacement of the ascending aorta was performed over the sinotubular junction. In 6 of patients, the indication was intraoperative detection of friable or destructured aortic wall; in the other 2, replacement was undertaken electively owing to progressive dilation detected on follow-up imaging after the index procedure. The prosthesis was supra-annular in all cases. Postoperative TEE showed trivial or no PVL in 98.5% of cases. No patch material was used for annular reconstruction.

Table 2.

Early (30-day) outcomes

Outcome	Total (N = 70)	Original (N = 47)	Expansion (N = 23)
Hospital mortality, n (%)	4 (5.7)	3 (6.4)	1 (4.3)
Stroke, n (%)	3 (4.3)	2 (4.3)	1 (4.3)
Dialysis, n (%)	6 (8.6)	4 (8.5)	2 (8.7)
Permanent pacemaker, n (%)	5 (7.1)	3 (6.4)	2 (8.7)
Ascending aorta replacement, n (%)	8 (11)	5 (11)	3 (13)

The prostheses implanted were Hancock and Avalus (Medtronic) prostheses for the bioprosthetic group and Bicarbon (Corcym) prostheses for the mechanical group. A total of 38 bioprosthetic valves were implanted (25 in the original group and 13 in the expansion group), and 32 mechanical prostheses were implanted (22 in the original group and 10 in the expansion group). The overall mean postoperative transvalvular gradient was 10.4 ± 3.1 mm Hg. The median implanted valve size was 23 mm (IQR, 21-25 mm).

In-hospital mortality was 5.7% (n = 4). Causes of death included refractory septic shock (n = 2), low-output syndrome (n = 1), and fulminant fungal sepsis (n = 1). Early neurologic events occurred in 3 patients (4.3%), dialysis was required in 6 patients (8.6%), and 5 patients (7.1%) received a permanent pacemaker. There were no cases of early prosthesis dysfunction or root dehiscence. Postoperative TEE confirmed unobstructed coronary flow in all patients.

Mid-Term Survival and Freedom from Reintervention

Follow-up was complete in all survivors (mean, 5.2 ± 1.3 years; median, 5.0 years; IQR, 4.1–6.3, years). At 5 years, overall survival was 86.0% (95% CI, 78.0%-93.9%). Freedom from the composite endpoint of PVE recurrence or aortic reoperation was 89.2% (95% CI, 82.1%-96.2%) (Figure 2). No significant differences in survival estimates were noted between the original and expansion groups (log-rank P = .71 and .64 for survival and event-free estimates, respectively).

Figure 2.

Kaplan-Meier curves of freedom from the composite endpoint. The estimated cumulative freedom from the composite endpoint at 1, 2, 3, 4, and 5 years was 93% (95% confidence interval [CI], 88%-97%) at 1 year, 93% (95% CI, 87%-97%) at 2 years, 88% (95% CI, 81%-93%) at 3 years, 88% (95% CI, 81%-93%) at 4 years, and 86% (95% CI, 78%-91%) at 5 years.

Two patients (2.9%) developed late reinfection (both treated medically), and 1 patient required reoperation for structural valve degeneration at 54 months. No cases of prosthetic dehiscence or root pseudoaneurysm occurred. The majority of patients (92%) remained in NYHA class I-II, and no new permanent pacemakers were required during follow-up.

At follow-up, no adverse events were attributed to the intentionally unobliterated subannular pouch: no PVL requiring intervention, no hemolysis, no pseudoaneurysm, and no imaging evidence of flow within the residual cavity on TEE or CT scan.

Echocardiographic Outcomes

The mean transvalvular gradient remained stable over time (10.4 ± 3.1 mm Hg at discharge vs 11.1 ± 3.6 mm Hg at 5 years; P = .27). Mean LVEF improved from 52 ± 10% at baseline to 58 ± 8% at final follow-up (P < .01). Mild PVL was observed in 2 patients (2.8%) on the last echocardiogram; none required intervention. There were no significant differences in hemodynamic parameters between mechanical and biological prostheses.

Predictors of Late Adverse Outcomes

On univariable analysis, age >75 years, LVEF <30%, serum creatinine >2 mg/dL, chronic renal failure, incomplete abscess excision, and fungal endocarditis were associated with the composite endpoint (Table 3). On multivariable analysis, age >75 years (hazard ratio [HR], 2.5; 95% CI, 1.0-6.2; P = .048), LVEF <30% (HR, 3.1; 95% CI, 1.2-8.1; P = .019), serum creatinine >2 mg/dL (HR, 2.8; 95% CI, 1.1-7.3; P = .029), chronic renal failure (HR, 2.6; 95% CI, 1.0-6.8; P = .045), and incomplete abscess excision (HR, 4.4; 95% CI, 1.3-14.8; P = .016) remained significant independent predictors of the composite endpoint. Fungal endocarditis and urgent/emergent surgery showed trends but did not reach statistical significance. The period of surgery (original vs expansion cohort) was not independently associated with outcomes (P = .71).

Table 3.

Univariable and multivariable Cox regression

Predictor	Univariable analysis	Multivariable analysis, HR (95% CI); P value
Age >75 y	Associated with composite endpoint	2.5 (1.0-6.2); .048
LVEF <30%	Associated with composite endpoint	3.1 (1.2-8.1); .019
Serum creatinine >2 mg/dL	Associated with composite endpoint	2.8 (1.1-7.3); .029
Chronic renal failure	Associated with composite endpoint	2.6 (1.0-6.8); .045
Urgent/emergent surgery	Associated with composite endpoint	2.2 (0.9-5.4); .08
Incomplete abscess excision	Associated with composite endpoint	4.4 (1.3-14.8); .016
Fungal endocarditis	Associated with composite endpoint	2.1 (0.8-5.6); .12
Period of surgery (original vs expansion cohort)	Not associated	Not significant; .71

HR, Hazard ratio; CI, confidence interval; LVEF, left ventricular ejection fraction.

Discussion

Comparison with Contemporary Surgical Outcomes

Our findings demonstrate encouraging mid-term outcomes with a standardized, patch-free supra-annular aortic valve replacement approach in patients with destructive PVE. At 5 years, overall survival reached 86%, and freedom from reinfection or reintervention was 89.2%, benchmarks that align favorably with contemporary series involving more complex root replacement strategies. For comparison, Yousif and colleagues⁷ reported a 10-year survival of 83% in a cohort undergoing cryopreserved homograft aortic root replacement for invasive PVE, emphasizing the importance of aggressive debridement and reconstruction of the infected root. Although root replacement may offer effective infection control, it is associated with longer ischemic times, greater technical complexity, and potential morbidity—particularly in frail or comorbid patients.

Importantly, our strategy avoids root replacement and instead relies on a standardized supra-annular, patch-free implantation technique. By eliminating the need for coronary reimplantation and annular patching, we significantly reduced mean cross-clamp time to 94 ± 28 minutes, compared to the >120 minutes typically reported in homograft series.⁸ This difference is clinically relevant; prolonged ischemic time is a well-established predictor of early postoperative mortality in endocarditis surgery. Al-Sarraf and colleagues⁹ identified both cross-clamp duration and preoperative renal dysfunction as independent predictors of long-term mortality, even among patients who survive the immediate postoperative phase.

In our cohort, although ischemic time reached statistical significance in univariate analysis, it did not retain independent prognostic value in the multivariate model. This may reflect the procedural homogeneity and limited variability in surgical approach, suggesting that simplification and standardization may attenuate the impact of classical risk factors in high-risk IE patients.

Both the study by Guihaire and colleagues¹⁰ and the long-term cohort reported by Amabile and colleagues¹¹ provide compelling evidence supporting the use of the Freestyle stentless xenograft in complex aortic valve surgeries. Guihaire and colleagues demonstrated the feasibility of excluding complex paraannular aortic abscesses using the Freestyle xenograft, highlighting its adaptability in challenging cases. Amabile and colleagues’ extensive prospective cohort of 500 patients further reinforces the durability and efficacy of the Freestyle bioprosthesis, reporting high rates of freedom from structural valve deterioration and excellent long-term cardiovascular survival. Importantly, both approaches are compatible with our patchless technique, as they do not require manipulation of the coronary ostia, making them suitable alternatives or complements in centers seeking to preserve coronary anatomy while addressing complex aortic root infection.

Technical Implications of a Patch-Free Supra-Annular Approach

One of the key technical advantages of our approach lies in the complete avoidance of patching, annular enlargement, or root replacement. The bioprosthesis is securely anchored in a supra-annular position directly onto debrided and structurally sound native tissue, with additional external felt reinforcement. This design allows for restoration of the annular continuity without introducing synthetic materials into the intracardiac lumen, thereby minimizing the prosthetic footprint within the infected field.

Compared to more complex reconstructions, such as cryopreserved homografts or full root replacement with coronary reimplantation, this simplified technique contributes to shorter cross-clamp and CPB times, which have been independently associated with improved outcomes in surgery.⁹ Moreover, by preserving annular geometry and avoiding bulky suture lines or patch material, our technique potentially reduces the local burden of prosthetic material—an important consideration in an infected and friable surgical field.

This aligns with the findings of Galeone and colleagues,¹² who emphasized that long-term success in aortic root endocarditis is more closely linked to complete infection clearance and stable annular reconstruction than to the specific type of valve conduit used, and that the use of homografts in patients with extensive annular abscess and multiple valve involvement is still safe despite the longer CPB and aortic cross-clamp times. Their observations support the rationale for limiting prosthetic contact with infected tissue whenever anatomically feasible. In contrast, Williams and colleagues¹³ reported no difference in mortality, reinfection, or reoperation rates between homografts and other valves or valved conduits in the management of complex aortic endocarditis.

Role of Annular Integrity and PVL in Long-Term Results

Preservation of the native annular architecture may have a crucial role in achieving optimal valve seating, sealing, and long-term prosthetic function. In our series, we observed no cases of moderate or severe PVL, supporting the mechanical stability and competence of supra-annular implantation on viable native tissue. In contrast, Huang and colleagues¹⁴ reported high mortality and a high incidence of PVL in patients with destruction of the aortic annulus. Zubarevich and colleagues⁵ reported a 30-day mortality of 8.9% and a PVL rate of 6.1% in a cohort of patients treated with sutureless aortic valves, noting particularly unfavorable outcomes in the subset with annular destruction. Their findings reflect the challenges of achieving adequate anchoring and sealing in disrupted anatomy when relying on radial-force-based prostheses.

Unlike sutureless or rapid-deployment devices, which are designed primarily for calcified but structurally intact annuli, conventional bioprostheses allow for tailored, circumferential suture placement on debrided tissue and reinforcement with external materials. This may result in improved hemostasis and reduced PVL risk in the context of fragile, infected surgical fields.

In anatomies with destructive IE characterized by aortomitral junction disconnection confined to the noncoronary sinus, long-term sealing can be achieved not by “maintaining annular integrity” in a circumferential sense, but rather by reconstructing a stable subcoronary anchoring plane at the noncoronary sinus, which thus acts as a functional neo-annulus. In this setting, a sutured bioprosthesis with external suture line buttressing limited to the noncoronary sinus showed durable sealing and satisfactory functional outcomes in our series. Conversely, when destruction is circumferential or coronary reimplantation is required, root replacement (Bentall/homograft) remains our preferred strategy.

Risk Stratification: Renal Function and Cross-Clamp Time

Preoperative renal dysfunction emerged as a significant predictor of late adverse events in our cohort, a finding that aligns with previous evidence highlighting impaired renal function as a key determinant of poor long-term outcomes in IE. In a large observational study, Al-Sarraf and colleagues⁹ demonstrated that both preoperative renal impairment and prolonged aortic cross-clamp duration were independently associated with increased long-term mortality in patients undergoing early surgery for IE.

In our series, while cross-clamp time was statistically significant in univariate analysis, it did not retain significance in the multivariable model. This may reflect the limited variability in surgical technique and procedural strategy, particularly the near-complete avoidance of root replacement. Our experience suggests that standardization and streamlining of surgical protocols when anatomically feasible may help attenuate the prognostic impact of operative duration in high-risk patients.

These observations underscore the importance of procedural discipline in complex IE surgery, particularly in frail patients with impaired baseline organ function, where minimizing surgical complexity and ischemic burden can yield tangible survival benefits.

Radical Debridement as the Cornerstone of Infection Control

Radical and complete debridement of infected tissue is universally recognized as the cornerstone of surgical success in PVE. In our cohort, incomplete debridement was independently associated with adverse outcomes (HR, 4.4; P = .016), highlighting the critical impact of achieving sterile margins even in the setting of severe tissue destruction.

While early surgery is known to improve outcomes in active IE, its benefit may be offset if residual infection remains.¹⁵ In our experience, a standardized supra-annular approach that prioritizes complete excision of infected and devitalized tissue without the need for root replacement proved effective in achieving durable infection control. Although Chu and colleagues,² on behalf of the International Collaboration on Endocarditis, demonstrated that early surgery reduces mortality in IE, they did not explicitly evaluate residual infection as a prognostic factor. However, surgical series have consistently emphasized that radical debridement is essential to minimize recurrence and ensure long-term success, particularly in patients with prosthetic material or annular invasion.

Limitations of Alternative Strategies

Homografts remain an essential tool in the surgical armamentarium for managing extensive aortic root destruction, particularly in cases involving PVE with abscess formation or annular disruption.¹⁶ However, their clinical use is limited by donor availability, technical complexity, and long-term durability concerns. Sutureless and rapid-deployment valves, although appealing for their technical simplicity and shorter operative times, appear to have limited utility in destructive PVE. Zubarevich and colleagues⁵ demonstrated that annular disruption significantly impairs anchoring of sutureless valves, resulting in higher rates of PVL (6.1%) and early failure. These devices rely on radial force for fixation and are optimized for use in calcified, intact annuli—conditions rarely present in endocarditis. In contrast, conventional bioprostheses, when securely sutured onto debrided and reinforced native tissue, offer a more predictable and adaptable strategy. Our results suggest that in selected patients with preserved annular architecture, this approach ensures effective infection control and long-term structural stability without the need for root replacement.

Limitations

This study has several limitations. First, its retrospective single-center design might have introduced selection and information bias, despite prospective data capture. Second, although our cohort of 70 patients is one of the largest single-technique series for destructive PVE, the sample size still limits the power of subgroup analyses and multivariable modeling, as reflected by wide confidence intervals. Third, a direct comparison with conventional aortic root replacement was not feasible in the present study, as no contemporaneous or historical control group was available. This represents an intrinsic limitation of our work, which prevents us from drawing definitive conclusions regarding the relative advantages of the proposed technique. Future larger, comparative, and, ideally, multicenter studies are needed to validate these findings. Fourth, we relied on echocardiographic assessment of valve function without systematic core lab adjudication or routine CT scans beyond 12 months, potentially underestimating subclinical degeneration. Finally, microbiological diversity and evolving antibiotic protocols over the 9-year study period could confound outcome comparisons. External validity is limited by the explicit anatomic prerequisites and contraindications adopted (eg, requirement for aortomitral junction disconnection; threatened or low-lying LM ostium), which may select anatomies amenable to secure subcoronary reconstruction.

Conclusions

In this expanded cohort of patients with PVE complicated by complex para-annular aortic abscess, the modified prosthetic root implantation technique demonstrated favorable outcomes while maintaining operative simplicity, graft availability, and ease of future coronary access. When considered alongside recent multicenter data, these findings suggest that the technique is a reliable, reproducible, and durable option for extensive but localized aortic PVE. The impact of preoperative risk factors and completeness of abscess excision highlights the importance of early surgical referral and thorough debridement. Further multi-institutional, propensity score–matched studies are needed to confirm these observations and to define the optimal reconstruction strategy for different patterns of infective destruction.

Conflict of Interest Statement

The authors reported no conflicts of interest.

The Journal policy requires editors and reviewers to disclose conflicts of interest and to decline handling or reviewing manuscripts for which they may have a conflict of interest. The editors and reviewers of this article have no conflicts of interest.