This study was reviewed and approved by the Ethics Committee of the Medical University of Vienna (Ethical Number: 1872/2022; approval date 16.01.2023). The individual patient's consent was waived.

Patients were screened using the institutional database of the Department of Cardiac Surgery at the Medical University of Vienna. Between January 2016 and December 2022, 220 consecutive patients requiring surgery for active IE were included in this retrospective study. Diagnosis of IE was based on the modified Duke criteria and current guidelines (3–5).

All patients underwent surgical valve replacement with either biological or mechanical prosthesis at the aortic or/and mitral position. The exclusion criteria were age < 18 years, aortic root replacement, homograft implantation, isolated mitral valve repair, and a combination of both suture-securing techniques.

The operations were performed by multiple senior surgeons. For all prostheses, the valve stitch placement technique involved the utilization of U-stitches with pledgets. The surgical treatment of IE at our institution is in line with current guidelines including radical debridement of all infected and necrotic tissue, followed by the reconstruction of any remaining defects (3). Additionally, during the surgery for IE all remaining tissue and surgical pledgets were soaked in an intraoperative disinfectant (povidone-iodine), which is standard practice at our department.

The patients were grouped based on the suture-securing technique used for prosthesis fixation (automated titanium fastener group vs. hand-tied knots group). Operation reports and institutional surgery protocols were reviewed to identify the suture-securing technique. The decision towards using either of the suture-securing techniques was based primarily on the surgeon's personal preference. The use of an automated titanium fastener device started in the year 2016 at our department.

The primary study endpoint was re-endocarditis during follow-up. The secondary study endpoints included stroke onset, all-cause mortality, and a composite outcome during follow-up. The composite outcome was defined as an occurrence of either re-endocarditis, stroke, or all-cause mortality during follow-up. A stroke was defined as an acute neurological deficit with evidence of a new lesion on a neuroimaging scan (6). The patient was considered to have had a stroke when either ischemic or hemorrhagic stroke was diagnosed. The study endpoints were identified from patients' medical records. Mortality data were obtained from the Austrian statistics department. No patient was lost to follow-up. A comprehensive dataset was generated through the prospectively maintained database at our department. The last date of follow-up was set for January 5, 2023. The median follow-up time was 664 days for the automated titanium fastener group and 900 days for the hand-tied knots group (p = 0.217).

Continuous variables are presented as median and 1st and 3rd quartile, and categorical variables as count and percentages. The Mann–Whitney U test was applied to analyze continuous variables with non-normal distributions. For the comparison of categorical variables, the chi-square test was used.

To estimate the effect of the suture-securing technique (automated titanium fastener group vs. hand-tied knots group) on the primary as well as secondary study endpoints, univariable and multivariable proportional hazards regression analyses were performed with the exposure time starting at the date of surgery. Based on directed acyclic graph using current published evidence and clinical experience we selected co-variables to be included in the multivariable models.

The analysis of the primary study endpoint (re-endocarditis) was performed using a multivariable proportional subhazards Fine and Gray regression model accounting for death as a competing event. The primary study endpoint was depicted by the Kaplan–Meier cumulative event curves of both study groups.

The secondary study endpoints: stroke and composite outcome were analyzed using multivariable Cox proportional hazards regression models. The effects were quantified as hazard ratio (HR) with 95% confidence interval (CI). The Kaplan–Meier cumulative event curves for the composite outcome were visualized for both study groups. For the secondary study endpoint, all-cause mortality, the multivariable Poisson regression model was used. The effects were calculated as an incidence-rate ratio with 95% CI.

We used the likelihood ratio test to assess deviations from linearity and to test for interactions.

For superiority, a two-sided p-value of less than 0.05 was considered to indicate statistical significance. SPSS 27.0 (IBM Corp, Armonk, NY, USA) and STATA 16.1 software (StataCorp LLC, College Station, TX, USA) were used for statistical analysis.

Suture-securing with an automated titanium fastener was performed in 114 (51.8%) patients, whereas the conventional technique of hand knot-tying was used in 106 (48.2%) patients (Table 1). The automated titanium fastener group had a significantly higher rate of preoperative atrial fibrillation (40.4% vs. 27.4%; p = 0.042) and diabetes mellitus (27.2% vs. 16.0%; p = 0.045). All other baseline characteristics were comparable between groups.

Table 2 lists species that were identified through blood culture before surgery. The rate of detected species was similar between study groups.

The operative characteristics are summarized in Table 3. No differences in eras were observed among study groups. Aortic valve replacement, mitral valve replacement, as well as combined aortic and mitral valve replacement were performed at similar rates. For aortic valve replacement, the Medtronic Avalus prosthesis was implanted significantly more frequently in the automated titanium fastener group (12.3% vs. 4.7%; p = 0.046) (Supplementary Table S1). The rate of annular patch augmentation, Commando procedure, ventricular septal defect closure, and aortic root enlargement was similar between groups (Table 3). Coronary artery bypass grafting (CABG) was performed with a significantly higher rate in the automated titanium fastener group (19.3% vs. 9.4%; p = 0.038). The cross-clamp times, as well as cardiopulmonary bypass (CPB) times, were comparable among groups.

During the follow-up period, 18 patients (8.2%) reached the primary study endpoint. Univariable Cox proportional hazards regression analysis (Table 4) demonstrated a significantly lower risk of re-endocarditis in the automated titanium fastener group (HR 0.35, 95% CI 0.12–0.96, p = 0.042). A multivariable proportional competing risk regression analysis showed that suture-securing with an automated titanium fastener was independently associated with a significantly lower risk of re-endocarditis during follow-up (adjusted sub-hazard ratio 0.33, 95% CI 0.11–0.99, p = 0.048) (Table 4; Supplementary Table S2). The median time to re-endocarditis was significantly longer in the automated titanium fastener group (271 days vs. 58 days p = 0.040). The Kaplan–Meier cumulative event curves of both study groups are depicted in Figure 2.

During the follow-up period, a stroke occurred in 36 patients (16.4%). The automated titanium fastener was not associated with an increased risk of stroke, as shown in the multivariable Cox proportional hazards regression model (adjusted HR 0.54, 95% CI 0.27–1.08, p = 0.082) (Table 4; Supplementary Table S3). Moreover, preoperative stroke (within 30 days) was independently associated with a significantly increased risk of stroke onset during the follow-up period (adjusted HR 2.14, 95% CI 1.02–4.45, p = 0.043) (Supplementary Table S3).

Mortality occurred in 58 patients (26.4%) during the follow-up period. As demonstrated in the multivariable Poisson regression analysis, the automated titanium fastener group was not associated with an increased risk of all-cause mortality (adjusted incidence-rate ratio 1.42, 95% CI 0.83–2.42, p = 0.202) (Table 4; Supplementary Table S4).

A composite outcome (defined as occurrence of either re-endocarditis, stroke, or all-cause mortality) was reached in 82 (37.3%) patients throughout the follow-up period. Univariable Cox proportional hazards regression analysis (Table 4) showed a significantly lower risk of experiencing the composite outcome in the automated titanium fastener group (HR 0.61, 95% CI 0.39–0.96, p = 0.031). In the multivariable Cox proportional hazards regression model, the automated titanium fastener was not associated with an increased risk of attaining the composite outcome (adjusted HR 0.65, 95% CI 0.42–1.02, p = 0.061) (Table 4; Supplementary Table S5). The Kaplan–Meier cumulative event curves were visualized for both study groups (Figure 3).

The rates of in-hospital postoperative adverse events were similar between the study groups (Table 5). Regarding outcomes, the rate of non-IE-related reoperation was significantly lower in the automated titanium fastener group (0.0% vs. 4.7%; p = 0.019) (Table 5). The causes of non-IE-related reoperation in the hand-tied knots group were major paravalvular leak (2.7%), aortic dissection (0.9%), ascending aortic aneurysm (0.9%), and ventricular aneurysm (0.9%). Noteworthy, the rate of re-endocarditis with an initial pathogen was significantly lower in the automated titanium fastener group (1.8% vs. 9.4%, p = 0.012).

No significant difference was observed regarding duration of antibiotic therapy between automated titanium fastener group and hand-tied knots group (51 days, interquartile range 42–66, vs. 58 days, interquartile range 43–78; p = 0.053) (Table 6).

The administration of erythrocyte concentrate, and thrombocyte concentrate were similar among the study groups during the intraoperative period and throughout the entire postoperative hospital stay, encompassing stay at the intensive care unit, intermediate care unit (if applicable), and standard care unit (Table 6).

The causes of death were comparable among the study groups (Table 7).

This study represents the first study to analyze long-term clinical outcomes based on the suturing technique used in prosthesis fixation for infective endocarditis, suggesting potentially lower risk of re-endocarditis when utilizing automated titanium fastener device.

Nonetheless, this study had some limitations: (1) It was retrospective single-center study with a relatively small study cohort; (2) Although the study groups were similar, they showed differences in terms of diabetes mellitus, preoperative atrial fibrillation, and CABG—concomitant procedure. (3) The occurrence of re-endocarditis is attributed to multiple factors, irrespective of the suture-securing technique being used; (4) The use of automated titanium fastener was depending on surgeon's preference; (5) The current literature on this topic is sparse.