Early outcomes and resource utilization across contemporary transcatheter and surgical valve interventions: a single-center observational study

Abstract

Objective

To evaluate early clinical outcomes and hospital resource utilization across three treatment strategies for valvular heart disease: median sternotomy, right mini-thoracotomy and transcatheter interventions.

Results

This retrospective single-center study included 109 consecutive adult patients undergoing valve interventions between January and October 2025 (sternotomy n = 20; mini-thoracotomy n = 25; transcatheter n = 64, including transcatheter aortic valve implantation and mitral/tricuspid edge-to-edge repair). Patients undergoing transcatheter procedures were older and had higher operative risk. Transcatheter interventions were associated with shorter ICU stay (median 1 [1–2] vs. 3 [2–4] days; p < 0.001) and hospital stay (5 [4–7] vs. 7 [5–10] and 9 [7–12] days; p < 0.001). In-hospital mortality registered one event, with no significant differences between groups (p = 1.00). Rates of stroke, acute kidney injury and permanent pacemaker implantation were low and similar between groups. Reoperation for bleeding occurred only in surgical patients, more frequently after sternotomy (10.0%) than mini-thoracotomy (4.0%) and was absent after transcatheter procedures (0.0%; p = 0.04). Similarly, the composite endpoint of in-hospital events did not differ significantly between groups (20.0% vs. 8.0% vs. 9.4%; p = 0.34).

Introduction

Valvular heart disease (VHD) remains a major cause of morbidity and mortality worldwide, with an increasing prevalence driven by population aging and improved survival of patients with cardiovascular comorbidities [1–6]. Over the past decades, the therapeutic landscape of VHD has undergone a profound transformation, evolving from conventional open-heart surgery performed through median sternotomy toward less invasive surgical approaches like ministernotomy and minithoracotomy and, more recently, towards transcatheter valve interventions [7–12]. This evolution reflects continuous efforts to reduce procedural trauma, perioperative complications and recovery time, while maintaining or improving clinical outcomes.

While median sternotomy has long represented the standard surgical approach due to its reproducibility and broad applicability, it is associated with greater surgical trauma and prolonged recovery [13–18]. Minimally invasive techniques, particularly right mini-thoracotomy, aim to reduce perioperative morbidity while preserving surgical efficacy. In parallel, transcatheter interventions such as transcatheter aortic valve implantation (TAVI) and edge-to-edge mitral and tricuspid repair have expanded therapeutic options, particularly for elderly or higher-risk patients [19–23]. While robust randomized trials have demonstrated the efficacy of these techniques in selected populations, real-world practice often encompasses a broader and more heterogeneous patient spectrum.

Most published studies compare two strategies at a time (e.g., surgery versus transcatheter therapy or sternotomy versus minimally invasive surgery), often within specific valve positions or risk categories. Direct comparisons across all three contemporary strategies within a single institutional framework remain limited, particularly in real-world settings where treatment allocation reflects multidisciplinary Heart Team decision-making rather than protocol-driven selection [24–28].

This study provides a real-world, single-center evaluation of early clinical outcomes and hospital resource utilization across these three treatment strategies.

The primary objectives of this study were:

(1) to compare early clinical outcomes among patients undergoing median sternotomy, minimally invasive surgery (right mini-thoracotomy) and transcatheter valve interventions;

(2) to assess differences in hospital resource utilization, with a focus on intensive care unit and total hospital length of stay;

(3) to provide a clinically oriented interpretation of the role of these treatment strategies in the context of contemporary valve management.

Materials and methods

Study design and ethical approval

This retrospective, observational, single-center cohort study included consecutive adult patients undergoing valve-related surgical or transcatheter interventions between January and October 2025 in a tertiary cardiovascular center. Clinical data were extracted from a prospectively designed institutional cardiovascular database (Epicard v5.2.2, Toulouse, France). All patients were screened chronologically to ensure consecutive inclusion and to minimize selection bias.

The study was approved by the Ethics Committee of Titu Maiorescu University, Bucharest, Romania (Approval No. 21/06.11.2025) and conducted in accordance with the Declaration of Helsinki and national regulations. All participants provided informed consent. Data were anonymized prior to analysis. As a non-interventional retrospective study, no prospective trial registration was required.

Patient selection

Inclusion criteria

Patients were eligible if they met all of the following criteria:

• Age ≥ 18 years

• Diagnosis of aortic, mitral or tricuspid valvular heart disease

• Intervention via:

  • Median sternotomy
  • Right mini-thoracotomy
  • Transcatheter approach (TAVI, MitraClip, TriClip or Pascal)

• Isolated valve procedures or valve procedures with minor concomitant interventions (e.g., patent foramen ovale or atrial septal defect closure)

Exclusion criteria

Patients were excluded for:

• Concomitant coronary artery bypass grafting

• Major aortic surgery beyond valve replacement

• Rare or highly heterogeneous congenital procedures (e.g., Ross procedure, Ebstein anomaly repair)

• Missing key variables required for classification or outcome analysis

Study groups

Patients were categorized into three predefined groups:

1. Median Sternotomy

2. Minimally Invasive Surgery (right mini-thoracotomy)

3. Transcatheter Intervention

Procedural techniques

Surgical procedures were performed under general anesthesia with cardiopulmonary bypass and standard myocardial protection according to institutional protocols and international guidelines. Valve repair included annuloplasty, leaflet reconstruction or chordal replacement. Valve replacement utilized mechanical or biological prostheses based on patient characteristics and shared decision-making.

Minimally invasive procedures were performed via right mini-thoracotomy with peripheral cannulation when required.

Transcatheter procedures were conducted in a hybrid operating room under fluoroscopic and transesophageal echocardiographic guidance. TAVI was performed using balloon-expandable or self-expanding valves via femoral or alternative vascular access when indicated. Mitral and tricuspid interventions employed edge-to-edge repair systems (MitraClip, TriClip or Pascal) following standard implantation protocols.

Treatment strategy was selected by the institutional Heart Team based on patient age, operative risk, comorbidity burden, anatomical suitability, valve pathology and procedural feasibility.

Statistical analysis

Statistical analyses were performed using IBM SPSS Statistics version 27.0 (IBM Corp., Armonk, NY, USA). Continuous variables are presented as mean ± standard deviation or median [interquartile range], as appropriate and were compared across groups using the Kruskal–Wallis test due to non-normal distribution. Categorical variables are presented as counts and percentages and were compared using Fisher’s exact test (with the Fisher–Freeman–Halton extension for contingency tables larger than 2 × 2), given the small sample size and low expected cell counts. All tests were two-sided, with p < 0.05 considered statistically significant.

Procedural characteristics were analyzed descriptively only, without formal statistical comparison, due to intrinsic differences between surgical and transcatheter treatment strategies.

The composite endpoint was defined based on clinically relevant in-hospital events, broadly aligned with contemporary Valve Academic Research Consortium-3 (VARC-3) criteria.

Given the observational nature of the study and heterogeneity of procedures, analyses were primarily descriptive and not intended to establish causal relationships between treatment strategies.

Results

Study population and group distribution

From the institutional valve intervention database, a total of 109 patients undergoing valve procedures during the study period were included in the final analysis. Patients were categorized into three main groups according to procedural strategy (Fig. 1):

Fig. 1.

Distribution of patients according to treatment strategy. data are presented as counts (percentages)

• Median sternotomy: 20 patients (18.3%)

• Mini-thoracotomy: 25 patients (22.9%)

• Transcatheter valve interventions (TAVI, MitraClip, TriClip, Pascal): 64 patients (58.7%)

Baseline demographic and clinical characteristics

The comparative analysis included patients undergoing valve interventions via median sternotomy, mini-thoracotomy and transcatheter approaches. Patients in the transcatheter group were significantly older than those in the surgical groups (77.8 ± 7.4 years vs. 66 ± 9 and 62 ± 7 years, p < 0.001). (Fig. 2).

Fig. 2.

Age comparison between surgical, mini-thoracotomy and transcatheter valve intervention groups with error representation. Data are presented as mean values ± SD, consistent with Table 1

The proportion of female patients did not differ significantly across groups (35.0% in sternotomy, 44.0% in mini-thoracotomy and 50.0% in transcatheter; p = 0.49).

Operative risk, assessed using EuroSCORE II, differed significantly between groups, with higher values observed in the transcatheter cohort (3.6 ± 1.6) compared to the surgical groups (2.28 ± 1.7 and 2.9 ± 1.2, p = 0.03). Hypertension was also more prevalent in the transcatheter group (76.6%) compared with sternotomy (55.0%) and mini-thoracotomy (48.0%) patients (p = 0.03).

No statistically significant differences were observed among groups with respect to diabetes mellitus (25.0%, 16.0% and 28.1%; p = 0.49), chronic kidney disease (20.0%, 8.0% and 12.5%; p = 0.71), cerebrovascular disease (15.0%, 8.0% and 9.4%; p = 0.77), hypercholesterolemia (65.0%, 72.0% and 73.4%; p = 0.92), peripheral arterial disease (20.0%, 24.0% and 20.3%; p = 0.83) or chronic obstructive pulmonary disease (5.0%, 8.0% and 9.4%; p = 0.82). (Table 1).

Table 1.

Baseline demographic and clinical characteristics of the study population stratified by treatment strategy

Variable	Median sternotomy n = 20	Mini-thoracotomy n = 25	Transcatheter n = 64	p-value
Age, years (mean ± SD)	66 ± 9	62 ± 7	77.8 ± 7.4	< 0.001
Female sex, n (%)	7 (35.0)	11 (44.0)	32 (50.0)	0.49
EuroSCORE II (mean ± SD)	2.28 ± 1.7	2.9 ± 1.2	3.6 ± 1.6	0.03
Hypertension, n (%)	11 (55.0)	12 (48.0)	49 (76.6)	0.03
Diabetes mellitus, n (%)	5 (25.0)	4 (16.0)	18 (28.1)	0.49
Chronic kidney disease, n (%)	4 (20.0)	2 (8.0)	8 (12.5)	0.71
Cerebrovascular disease, n (%)	3 (15.0)	2 (8.0)	6 (9.4)	0.77
Hypercholesterolemia, n (%)	13 (65.0)	18 (72.0)	47 (73.4)	0.92
Peripheral arterial disease, n (%)	4 (20.0)	6 (24.0)	13 (20.3)	0.83
Chronic obstructive pulmonary disease, n (%)	1 (5.0)	2 (8.0)	6 (9.4)	0.82

Continuous variables are presented as mean ± standard deviation and were compared across groups using the Kruskal–Wallis test due to non-normal distribution. Categorical variables are presented as counts and percentages and were compared using Fisher’s exact test (with the Fisher–Freeman–Halton extension for contingency tables larger than 2 × 2). All tests were two-sided, with p < 0.05 considered statistically significant

Procedural characteristics

From the institutional valve intervention database, a total of 109 patients undergoing valve procedures during the study period were included in the final analysis. Patients were categorized into three groups according to procedural strategy: median sternotomy (n = 20), mini-thoracotomy (n = 25) and transcatheter valve interventions (n = 64). (Table 2)

Table 2.

Summary of procedural characteristics, valve position and intervention type stratified by treatment strategy

Variable	Sternotomy (n = 20)	Mini-thoracotomy (n = 25)	Transcatheter (n = 64)
Valve position
— Aortic	13 (65.0)	15 (60.0)	56 (87.5)
— Mitral	6 (30.0)	7 (28.0)	6 (9.4)
— Tricuspid	1 (5.0)	3 (12.0)	2 (3.1)
Procedure type
— Valve replacement	12 (60.0)	17 (68.0)	NA
— Valve repair	8 (40.0)	8 (32.0)	NA
Prosthesis type (surgical)
— Mechanical	8 (40.0)	3 (12.0)	NA
— Biological	4 (20.0)	14 (56.0)	NA
Transcatheter procedure type
— TAVI	—	—	56 (87.5)
— MitraClip / Pascal	—	—	6 (9.4)
— TriClip / Pascal	—	—	2 (3.1)
Alternative vascular access	NA	NA	10 (15.6)

Data are presented as counts and percentages. This table is descriptive in nature and no statistical comparisons were performed due to intrinsic differences between surgical and transcatheter treatment strategies and the lack of direct comparability across all variables

In the sternotomy group (n = 20), aortic valve interventions were the most frequent (65.0%), followed by mitral (30.0%) and tricuspid procedures (5.0%). Valve replacement was the predominant strategy (60.0%), while valve repair was performed in 40.0% of cases. Among patients undergoing valve replacement, mechanical prostheses were more frequently used than biological prostheses (40.0% vs. 20.0%).

In the mini-thoracotomy group (n = 25), aortic valve interventions were also the most common (60.0%), followed by mitral (28.0%) and tricuspid procedures (12.0%). Valve replacement remained the main approach (68.0%), although valve repair was performed in 32.0% of patients. In contrast to the sternotomy group, biological prostheses were more frequently used than mechanical valves (56.0% vs. 12.0%). (Fig. 3)

Fig. 3.

Distribution of valve position according to treatment strategy. Comparison of aortic, mitral and tricuspid valve involvement across sternotomy, mini-thoracotomy and transcatheter groups. Data are presented as counts (n)

In the transcatheter group (n = 64), procedures were predominantly performed on the aortic valve (87.5%), with fewer interventions targeting the mitral (9.4%) and tricuspid valves (3.1%). Transcatheter aortic valve implantation (TAVI) was the dominant modality (87.5%), followed by mitral (MitraClip/Pascal, 9.4%) and tricuspid interventions (TriClip/Pascal, 3.1%). Alternative vascular access was required in 15.6% (n = 10) of transcatheter procedures.

Overall, both surgical approaches were characterized by a predominance of aortic valve interventions and valve replacement strategies. In contrast, transcatheter procedures were almost exclusively focused on the aortic valve, with TAVI representing the most frequently performed intervention. While aortic valve disease predominated across all groups, mitral and tricuspid pathologies were more commonly managed surgically. (Table 2)

Early clinical outcomes

Patients undergoing transcatheter valve interventions experienced significantly shorter intensive care unit stays compared with both surgical approaches (median 1 [1–2] days vs. 3 [2–4] days in both sternotomy and mini-thoracotomy groups, p < 0.001). Similarly, overall hospital length of stay differed significantly across treatment strategies, with the shortest duration observed in the transcatheter group (median 5 [4–7] days), followed by mini-thoracotomy (7 [5–10] days) and sternotomy (9 [7–12] days) (p < 0.001) (Table 3; Fig. 4).

Table 3.

In-hospital and early clinical outcomes stratified by treatment strategy

Outcome	Sternotomy (n = 20)	Mini-thoracotomy (n = 25)	Transcatheter (n = 64)	p-value
In-hospital mortality, n (%)	0 (0.0)	0 (0.0)	1 (1.6)	1.00
ICU stay, days	3 (2–4)	3 (2–4)	1 (1–2)	< 0.001
Hospital stay, days	9 (7–12)	7 (5–10)	5 (4–7)	< 0.001
Reoperation for bleeding, n (%)	2 (10.0)	1 (4.0)	0 (0.0)	0.04
Stroke, n (%)	1 (5.0)	0 (0.0)	1 (1.6)	0.84
Acute kidney injury, n (%)	0 (0.0)	1 (4.0)	2 (3.1)	0.92
Permanent pacemaker implantation, n (%)	1 (5.0)	0 (0.0)	3 (4.7)	0.48
Composite endpoint (death, stroke, AKI, PPM, reoperation), n (%)	4 (20.0)	2 (8.0)	6 (9.4)	0.34

Continuous variables (ICU stay and hospital stay) are presented as median [interquartile range] and were compared across groups using the Kruskal–Wallis test due to non-normal distribution. Categorical variables are presented as counts and percentages and were compared using Fisher’s exact test (Fisher–Freeman–Halton extension for contingency tables larger than 2 × 2), given the small sample size and low expected cell counts. All tests were two-sided, with p < 0.05 considered statistically significant

Fig. 4.

Distribution of in-hospital and early postoperative outcomes according to treatment strategy (median sternotomy, mini-thoracotomy and transcatheter interventions). Data are presented as median [interquartile range] for continuous variables and as counts (percentages) for categorical variables

In-hospital mortality was extremely low (one event overall), with no statistically significant difference between groups (p = 1.00). However, meaningful comparison is limited due to the very low event rate. No statistically significant differences were observed with respect to the incidence of stroke (5.0% vs. 0.0% vs. 1.6%, p = 0.84), permanent pacemaker implantation (5.0% vs. 0.0% vs. 4.7%, p = 0.48) or acute kidney injury (0.0% vs. 4.0% vs. 3.1%, p = 0.92).

Reoperation for bleeding occurred only in surgically treated patients, being more frequent in the sternotomy group (10.0%) compared with mini-thoracotomy (4.0%) and absent in the transcatheter group (0.0%), reaching statistical significance (p = 0.04) (Table 3; Fig. 4).

To provide an integrated assessment of clinically relevant adverse events, a composite endpoint broadly aligned with contemporary VARC-3 concepts was constructed including in-hospital mortality, stroke, acute kidney injury, permanent pacemaker implantation and reoperation for bleeding. The composite endpoint occurred in 20.0% of patients in the sternotomy group, 8.0% in the mini-thoracotomy group and 9.4% in the transcatheter group, without statistically significant differences between groups (p = 0.34) (Table 3).

Discussion

In this real-world, single-center cohort, we compared contemporary surgical (median sternotomy and mini-thoracotomy) and transcatheter valve interventions with respect to early clinical outcomes, hospital resource utilization and their role within current treatment strategies.

No statistically significant differences in early in-hospital outcomes were observed across the three groups; however, these findings should be interpreted in the context of limited statistical power and baseline differences between cohorts. Our results are consistent with findings from major randomized trials and large international registries evaluating surgical versus transcatheter valve interventions [29–34]. Major randomized trials such as PARTNER and SURTAVI demonstrated non-inferiority of transcatheter approaches compared with surgery in terms of early mortality and stroke, even in higher-risk populations [35–39]. Reported mortality rates in these trials (approximately 4–6% at 1 year) and stroke rates (~ 2–3%) are comparable to those observed in our cohort, suggesting consistency of our observations with previously reported data. Importantly, the absence of significant differences in major complications in our study—including stroke, acute kidney injury and mortality—is also aligned with contemporary meta-analyses showing similar rates between transcatheter and surgical approaches, despite differences in patient risk profiles [40–44].

When assessed using a VARC-3–aligned composite endpoint, no statistically significant differences in the composite endpoint were observed across groups, supporting the feasibility and short-term safety of both strategies within this cohort [45, 46]. This approach enhances the comparability of our findings with contemporary valvular intervention studies.

A key finding of our study is the significantly reduced ICU and hospital length of stay associated with transcatheter interventions. In our cohort, median ICU stay was 1 day for transcatheter procedures versus 3 days for surgery and hospital stay was reduced to 5 days compared with 7–9 days for surgical approaches. These results closely mirror previously reported data, where ICU stay after transcatheter procedures is typically around 1–2 days compared with 3–4 days for surgery and total hospitalization approximately 4–5 days versus 9–10 days [40–44]. This consistent reduction reflects the lower invasiveness of transcatheter techniques and has important implications for healthcare resource utilization.

Procedure-specific trade-offs observed in our study are also in line with existing literature. Reoperation for bleeding occurred exclusively in surgical patients, whereas permanent pacemaker implantation was numerically higher in the transcatheter group. This pattern has been consistently reported, with studies showing lower bleeding rates after transcatheter procedures (approximately 5% vs. 9–10% for surgery) but higher pacemaker implantation rates (up to 10–12% vs. 3–5%) [47–50]. These differences highlight the importance of individualized risk–benefit assessment when selecting the optimal treatment strategy. Within the surgical cohort, minimally invasive mini-thoracotomy was associated with younger patients and lower operative risk compared with sternotomy, while maintaining similar clinical outcomes and shorter hospital stay. This supports its role as an intermediate strategy between full sternotomy and transcatheter intervention, combining minimal invasiveness with the durability of surgical repair or replacement.

From a guideline perspective, our findings reflect current recommendations from the ESC/EACTS guidelines, which emphasize the importance of patient selection based on age, surgical risk, comorbidities and anatomical considerations, rather than a one-size-fits-all approach [51–53]. Transcatheter interventions are generally recommended in older or higher-risk patients, while surgery remains preferred in younger or lower-risk individuals with longer life expectancy [20, 22, 25, 28, 31, 32]. The distribution of patients in our cohort closely mirrors these real-world decision pathways.

Finally, all procedures were performed by a single, integrated team experienced in both surgical and transcatheter techniques. This unified approach may reduce variability related to operator experience and reflects a Heart Team–based strategy in which treatment selection is guided by patient characteristics and clinical indication.

Study limitations

This study has several limitations. Its retrospective, single-center design limits generalizability and introduces potential selection bias, as transcatheter interventions were preferentially performed in older and higher-risk patients. In addition, no formal adjustment using propensity score methods or multivariable analysis was performed; therefore, residual confounding cannot be excluded and comparisons between treatment strategies should be interpreted as descriptive rather than causal.

The relatively small sample size and low event rates for key outcomes, including mortality, stroke and permanent pacemaker implantation, further limit statistical power and increase the risk of type II error, precluding definitive conclusions regarding less frequent complications. Moreover, the inclusion of heterogeneous valve pathologies (aortic, mitral and tricuspid) reflects real-world practice but introduces clinical variability that may influence outcomes and limit direct comparability between groups. Importantly, the heterogeneity of valve pathologies and procedures limits direct comparability between treatment strategies and findings should be interpreted as descriptive rather than indicative of treatment superiority.

Finally, only early in-hospital outcomes were evaluated, without assessment of long-term endpoints such as valve durability, reintervention rates, or survival. Treatment allocation was not randomized but based on Heart Team decision-making, which, although representative of contemporary clinical practice, may further contribute to selection bias. Future multicenter studies with larger sample sizes and extended follow-up are warranted to confirm these findings and better define long-term comparative effectiveness.

Conclusion

In contemporary clinical practice, optimal valve intervention requires an individualized balance between procedural invasiveness, patient risk profile and anticipated recovery. In this real-world cohort, transcatheter valve interventions were associated with no statistically significant differences in early clinical outcomes compared with surgical approaches, while offering advantages in terms of shorter intensive care unit and overall hospital length of stay. These findings support the use of transcatheter therapies as a clinically relevant option for elderly patients and those with increased comorbidity burden, in whom rapid recovery and reduced procedural stress are key considerations.

Surgical valve interventions, performed via median sternotomy or minimally invasive mini-thoracotomy, remain essential for patients requiring durable valve solutions, complex reparative strategies or comprehensive anatomical correction. The higher incidence of reoperation for bleeding observed after surgical procedures highlights an important procedure-specific consideration that should be incorporated into patient selection and preprocedural counseling.

Importantly, the integration of surgical and transcatheter techniques within a single experienced Heart Team enables a patient-tailored approach to valvular heart disease management and is associated with low in-hospital mortality in this cohort. Overall, these findings support a tailored, patient-centered approach to valvular heart disease management, in which surgical and transcatheter strategies are integrated as complementary options based on individual patient characteristics, in line with current guideline recommendations.

Abbreviations

AKI

Acute kidney injury

ICU

Intensive care unit

LOS

Length of stay

MS

Median sternotomy

MICS

Minimally invasive cardiac surgery

PPM

Permanent pacemaker

TAVI

Transcatheter aortic valve implantation

TEER

Transcatheter edge-to-edge repair

VHD

Valvular heart disease

Author contributions

Conceptualization, A.D.M., A.C. and V.S.C.; methodology, A.D.M., A.C., S.A.O. and V.S.C.; validation, A.C. and V.S.C.; formal analysis, A.D.M., A.C., S.A.O. and V.S.C.; investigation, A.D.M., A.C., S.A.O. and V.S.C.; data curation, A.D.M., A.C., S.A.O. and V.S.C.; writing—original draft preparation, A.D.M., A.C. and S.A.O.; writing—review and editing, A.D.M., A.C., S.A.O. and V.S.C.; supervision, V.S.C. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data availability

All relevant data supporting the findings of this study are available within the article. Additional datasets generated and analyzed during the current study are not publicly available due to institutional and patient privacy restrictions, but are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

The study was approved by the Ethics Committee of Titu Maiorescu University, Bucharest, Romania (Approval No. 21/06.11.2025) and was conducted in accordance with the Declaration of Helsinki.

Informed consent

All participants provided informed consent and all patient data were fully anonymized prior to analysis and analyzed in accordance with institutional regulations.

Competing interests

The authors declare no competing interests.