Abstract
Background
Minimally invasive mitral valve surgery (MIMVS) has become the standard procedure for treating mitral valve pathologies. However, the existing cardiac risk model fails to consider the distinctive perfusion and ventilation techniques of MIMVS, leading to inaccurate prediction of perioperative risks. This study aimed to identify the perioperative risk factors for major adverse cardiovascular events (MACEs) in MIMVS and develop a predictive model based on these factors.
Methods
This single-center retrospective study recruited 480 patients undergoing MIMVS at Beijing Anzhen Hospital between April 2010 and May 2024 and collected data on 79 perioperative clinical variables. The primary outcome was MACE within 30 days postoperatively. Univariate Cox regression analysis was used to analyze the associations between variables and outcomes, whereas elastic net regression was used to develop a risk prediction model (CompliMit Score) for MACE. The model was validated using 200 bootstrap replicates.
Results
The 30-day MACE rate was 12%, and 31 clinical variables significantly correlated with MACE: 13 preoperatively, 9 intraoperatively, and 9 postoperatively. From these, we developed the CompliMit Score, which included 14 risk factors identified through elastic net regression. The CompliMit Score identified more high-risk patients for MACE than the European System for Cardiac Operative Risk Evaluation II {area under the curve: 0.92 [95% confidence interval (CI): 0.88–0.96] vs. 0.67 (95% CI: 0.59–0.75)}, and internal validation confirmed its superior predictive performance.
Conclusions
Factors influencing MIMVS prognosis included preoperative, intraoperative, and postoperative variables. The newly developed CompliMit Score effectively identified patients who are at high risk of perioperative MACE, thus facilitating targeted postoperative care and resource allocation.
Keywords: Minimally invasive mitral valve surgery (MIMVS), mitral valve surgery, major adverse cardiovascular events (MACEs), risk prediction model
Highlight box.
Key findings
• The CompliMit Score effectively identified patients who are at high risk of perioperative major adverse cardiovascular events (MACEs), thus facilitating targeted postoperative care and resource allocation.
What is known and what is new?
• The American Society of Thoracic Surgeons risk score and European System for Cardiac Operative Risk Evaluation II (EuroSCORE II) do not adequately consider the unique, distinctive aspects of minimally invasive mitral valve surgery (MIMVS).
• We developed a predictive model, namely, the CompliMit Score, to stratify risk in patients undergoing MIMVS.
What is the implication, and what should change now?
• The findings of this study offer a framework of clinical variables that clinicians should pay special attention to during the perioperative period. This framework can be used to aid in the early identification of high-risk patients with MACE and the development of more suitable treatment strategies. Future multicenter studies are required to validate the CompliMit Score across diverse populations and surgical procedures.
Introduction
After approximately 20 years, minimally invasive mitral valve surgery (MIMVS) via a right chest rib incision has become the standard procedure for mitral valve disorders (1,2). MIMVS is performed through smaller intercostal incisions, providing better cosmetic results, lower transfusion rates, faster perioperative recovery, and shorter hospital stays than traditional median sternotomy surgery (3-5). The endoscopic assistance employed in MIMVS enhances visualisation for the operating surgeon and the surgical team (6). Notably, modern thoracoscopic systems now provide image resolution sufficient to facilitate intricate valve reconstructions that previously required open exposure (7). Nevertheless, persistent concerns among clinicians and patients regarding the safety-efficacy balance of minimally invasive approaches for complex mitral interventions continue to drive academic debate, despite accumulating evidence supporting procedural viability.
The American Society of Thoracic Surgeons risk score and European System for Cardiac Operative Risk Evaluation (EuroSCORE) II score, which are currently widely used, are mainly derived from data on traditional surgeries and do not adequately consider the unique, distinctive aspects of MIMVS, such as variations in perfusion and ventilation methods (8). These systems primarily assess the impact of preoperative conditions and surgical methods on mortality; however, the surgical procedure also plays a crucial role in mortality. Therefore, to accurately predict the risk of MIMVS, it is essential to consider the risk factors before and during surgery and to evaluate the early postoperative status of patients.
This study aimed to identify the surgical risk factors predicting perioperative complications in MIMVS. Furthermore, we aimed to develop a predictive model to stratify risk in patients undergoing MIMVS. This approach provides patients with a more personalized and accurate assessment of surgical risks. To prioritize safety, we aimed to ensure that a greater number of patients would benefit from MIMVS. We present this article in accordance with the TRIPOD + AI reporting checklist (available at https://cdt.amegroups.com/article/view/10.21037/cdt-2025-25/rc).
Methods
Study population
In this single-center study, we recruited 480 patients who underwent MIMVS at Beijing Anzhen Hospital between April 2010 and May 2024. The diagnosis for all patients included moderate or severe pathological changes of the mitral valve, prosthetic valve dysfunction, or vegetation formation as determined by transthoracic echocardiography. Figure 1 shows the patient flow chart and admittance standards of the study.
Figure 1.
Enrollment flowchart for patients included in the final analysis. MACE, major adverse cardiovascular events; MIMVS, minimally invasive mitral valve surgery.
The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This retrospective analysis received approval from the Ethics Committee of Beijing Anzhen Hospital (No. KS2023092) with a waiver of individual informed consent. This study protocol was prospectively registered at the Chinese Clinical Trial Registry (registration No. ChiCTR2300079282).
Clinical data collection
Clinical data, including information from the electronic medical record system, imaging data, surgical management system, and outpatient follow-up system, were obtained from the medical database of Beijing Anzhen Hospital. Data collection included demographic characteristics, medical history, vital signs, surgery-related features, and hematological and biochemical indicators. In addition, we conducted frequent arterial blood gas analyses and chest radiography within 24 h postoperatively. A total of 79 clinical variables were collected, with 51, 11, and 17 variables gathered before, during, and after surgery, respectively.
Surgical technique
The procedural and technical details of MIMVS are well documented (9). Preparation involved double-lumen intubation for lung isolation, radial artery cannulation for continuous pressure monitoring, and venous access via the right internal jugular vein. Intraoperatively, transesophageal echocardiography was routinely performed to guide the evaluation of diseased valves and assess postoperative outcomes. Surgery was performed through an incision in the fourth intercostal space near the midaxillary line, regardless of whether the tricuspid valve was also being treated. Myocardial protection was achieved through antegrade perfusion of cold cardioplegia. To prevent embolism formation, CO2 was continuously injected into the right thoracic cavity while opening the heart cavity. The mitral valve was accessed via an incision in the interatrial groove. However, when combined with tricuspid valve surgery, an additional right atrial incision was required.
Definitions of events and end points
The primary endpoint was a composite of major adverse cardiovascular events (MACE) 30 days postoperatively. MACE was defined as low cardiac output syndrome, bleeding complications, renal dysfunction, stroke and other cerebrovascular events, and pulmonary and vascular complications. By combining the academic research alliances of the Valve Academic Research Consortium and the Mitral Valve Academic Research Consortium in the assessment of surgical safety, we reaffirmed the definitions of these six complications (10-12).
Low cardiac output syndrome: cardiac index ≤2.0 L·min−1·m−2. Mechanical assistance devices are necessary to maintain a systolic blood pressure of >90 mmHg.
Bleeding complications: drainage exceeds 1,000 mL within 3 h postoperatively, or re-exploration of the chest may be required to check for potential active bleeding. Clinically significant bleeding includes gastrointestinal, pulmonary, and cerebral bleeding.
Renal dysfunction: acute kidney injury or progression of chronic kidney disease. It is characterized by a two-fold increase in postoperative creatinine levels compared to preoperative levels and/or a urine output of less than 0.5 mL·kg−1·h−1 persisting for 12 h or more. It also involves undergoing continuous blood purification therapy.
Stroke and other cerebrovascular events: postoperative epilepsy or central nervous system dysfunction lasting for a period exceeding 72 h in clinical settings. Abnormalities in the nervous system, such as a new hemorrhage, stroke, or severe brain edema, are evident in imaging studies.
Pulmonary complications: postoperative acute respiratory distress with oxygen saturation (SpO2) ≤97% and SpO2/fraction of inspiration oxygen (FiO2) ≤315 mmHg, accompanied by radiographic evidence of pulmonary edema, or treatment includes repeat intubation or tracheotomy.
Vascular complications: direct blood vessel damage, thrombus formation, or retrograde dissection caused by the insertion of a peripheral vascular catheter.
Statistical analyses
The Kolmogorov-Smirnov test was used to determine the distribution of continuous variables. Continuous variables with normal distributions were presented as mean and standard deviation (SD) and were analyzed using a t-test. Otherwise, variables were presented as median [interquartile range (IQR)] and were analyzed using the Mann-Whitney U test. Categorical variables were presented as n (%) and were analyzed using Fisher’s exact or the Chi-squared test. The univariate Cox regression model was used to evaluate the correlation between clinical variables and MACE. The effect size of Cox regression analysis is expressed as the crude hazard ratio (Crude HR) with its corresponding 95% confidence interval (CI). A two-tailed test was used, and statistical significance was set as P<0.05. Baseline characteristics were analyzed using SPSS 29 (IBM, Armonk, NY, USA). All other calculations were performed using R 3.4.1 (The R Foundation, Vienna, Austria).
Risk prediction model choice, tuning, and testing
An elastic net regression model was used to construct a predictive model for MACE. A flowchart of the development and validation process of the CompliMit Score is shown in Figure 2. This model integrates the L1 penalty of least absolute shrinkage and selection operator (LASSO) and the L2 penalty of Ridge to address multicollinearity and reduce the complexity of high-dimensional data while allowing for variable selection and regularization. To develop the model, we scaled the data and analyzed 29 clinical variables related to MACE using logistic regression with an elastic net penalty. To determine the optimal parameters, we used 10-fold cross-validation and tested 10 different regularization strengths (λ). Internal validation was performed using 200 bootstrap replicates. Receiver operating characteristic (ROC) curve analysis was conducted to assess the predictive ability of the newly developed model and EuroSCORE II model for 30-day MACE. The calibration of the model was evaluated by examining the intercept and slope of the calibration curve to demonstrate the relationship between the observed 30-day MACE occurrence rates and the predicted rates. We used decision curve analysis to compare the clinical utility of the CompliMit Score and EuroSCORE II, and the ability of the model to stratify MACE risk was evaluated using Kaplan-Meier curves with statistical testing conducted using the DeLong method.
Figure 2.
Flowchart of the development and validation process of the CompliMit Score. MIMVS, minimally invasive mitral valve surgery.
Results
Baseline characteristics
This study enrolled 480 consecutive patients who underwent MIMVS. The included patients were categorized into the MACE (n=58) and No MACE (n=422) groups based on MACE occurrence. Table 1 presents the baseline characteristics of the two groups. The average age of patients was significantly higher in the MACE group than in the No MACE group (60.5±11.7 vs. 55.2±12.1 years, P<0.001). Regarding medical history, patients in the MACE group who underwent percutaneous coronary intervention (PCI) had a higher incidence (15.5% vs. 4.7%, P=0.004) and were more likely to have chronic liver disease (10.3% vs. 2.1%, P=0.005). Severe pulmonary hypertension was more common in the MACE group than in the No MACE group (10.3% vs. 3.6%, P=0.001). Furthermore, the incidence of severe tricuspid regurgitation was significantly higher in the MACE group than in the No MACE group (31.0% vs. 13.0%, P<0.001).
Table 1. Demographics and baseline clinical data.
| Characteristic | No MACE (N=422) | MACE (N=58) | P |
|---|---|---|---|
| Age (years) | 55.16±12.10 | 60.52±11.73 | <0.001 |
| Gender | 0.06 | ||
| Male | 161 (38.2) | 28 (48.3) | |
| Female | 261 (61.8) | 30 (51.7) | |
| BMI (kg/m2) | 23.69 (21.48–26.17) | 23.18 (19.97–25.31) | 0.08 |
| Smoking history | 79 (18.7) | 14 (24.1) | 0.38 |
| Drinking history | 46 (10.9) | 9 (15.5) | 0.38 |
| Hypertension | 106 (25.1) | 17 (29.3) | 0.52 |
| Hyperlipidemia | 22 (5.2) | 4 (6.9) | 0.76 |
| Diabetes | 34 (8.1) | 4 (6.9) | 0.81 |
| Chronic lung disease | 31 (7.3) | 7 (12.1) | 0.30 |
| Chronic liver disease | 9 (2.1) | 6 (10.3) | 0.005 |
| Chronic kidney disease | 7 (1.7) | 2 (3.4) | 0.61 |
| Neurological disorder | 33 (7.8) | 5 (8.6) | >0.99 |
| Poor mobility | 10 (2.4) | 1 (1.7) | >0.99 |
| Extracardiac arteriopathy | 163 (38.6) | 31 (53.4) | 0.04 |
| Atrial fibrillation | 227 (53.8) | 28 (48.3) | 0.48 |
| Previous MI | 4 (0.9) | 3 (5.2) | 0.04 |
| Previous PCI | 20 (4.7) | 9 (15.5) | 0.004 |
| PAH | 0.001 | ||
| Normal | 145 (34.4) | 11 (19) | |
| Mild | 200 (47.4) | 28 (48.3) | |
| Moderate | 62 (14.7) | 13 (22.4) | |
| Severe | 15 (3.6) | 6 (10.3) | |
| Previous cardiac surgery | 61 (14.5) | 17 (29.3) | 0.006 |
| NYHA | 0.17 | ||
| I | 2 (0.5) | 1 (1.7) | |
| II | 203 (48.1) | 22 (37.9) | |
| III | 175 (41.5) | 26 (44.8) | |
| IV | 42 (10) | 9 (15.5) | |
| Mitral valve pathology* | 0.13 | ||
| Stenosis | 116 (27.4) | 11 (19.0) | |
| Regurgitation | 181 (42.9) | 21 (36.2) | |
| Stenosis and regurgitation | 64 (15.2) | 9 (15.5) | |
| Tricuspid regurgitation | <0.001 | ||
| Mild | 284 (67.3) | 26 (44.8) | |
| Moderate | 83 (19.7) | 14 (24.1) | |
| Severe | 55 (13.0) | 18 (31.0) | |
| LVEF <50% | 15 (3.6) | 2 (3.4) | 0.82 |
| Procedures | <0.001 | ||
| Isolated | 227 (53.8) | 19 (32.8) | |
| Concomitant procedures | 195 (46.2) | 39 (67.2) |
Values are presented as numbers (percentages), mean ± standard deviation, or median (interquartile range). *, the totals for “Mitral Valve Pathology” exclude patients undergoing surgery for prosthetic valve dysfunction, infective endocarditis with vegetation formation. Stenosis/regurgitation categories are restricted to primary pathologies to avoid overlap. BMI, body mass index; LVEF, left ventricular ejection fraction; MACE, major adverse cardiovascular events; MI, myocardial infarction; NYHA, New York heart association; PAH, pulmonary arterial hypertension; PCI, percutaneous coronary intervention.
Procedural characteristics and outcomes
Regarding surgical characteristics, the MACE group had a higher proportion of patients undergoing two or more major cardiac surgeries (67.2% vs. 46.2%, P<0.005), longer surgical time (6.0±2.6 vs. 4.6±1.1 h, P<0.001), and longer cardiopulmonary bypass (CPB) time (183.5±121.5 vs. 134.5±42.4 min, P<0.001). Intraoperative bleeding (800 vs. 600 mL, P=0.005) and transfusion requirements (7.5 vs. 0 U, P<0.001) were also higher. However, no statistically significant differences were found in the types and sizes of implantations, intraoperative urine output, or aortic cross-clamping time (P>0.005). Table 2 summarizes the surgery-related characteristics of the two groups.
Table 2. Cardiac operative data.
| Characteristic | No MACE (N=422) | MACE (N=58) | P |
|---|---|---|---|
| Surgery duration (hours) | 4.6±1.1 | 6.0±2.6 | <0.001 |
| Implantations | 0.15 | ||
| Annuloplasty ring | 77 (18.2) | 10 (17.2) | |
| Biological valve | 158 (37.4) | 26 (44.8) | |
| Mechanical valve | 173 (41.0) | 20 (34.5) | |
| Others | 14 (3.3) | 2 (3.4) | |
| Surgical variables | |||
| Urine output (mL) | 1,028.3±640.6 | 1,210.3±878.0 | 0.40 |
| Blood loss (mL) | 600 [500–800] | 800 [500–1,000] | 0.005 |
| Aortic cross-clamping time (min) | 81.6±40.2 | 87.8±56.6 | 0.20 |
| CPB time (min) | 134.5±42.4 | 183.5±121.5 | <0.001 |
| 24-hour drain output (mL) | 398.1±313.0 | 785.6±669.3 | <0.001 |
| RBC transfusion (U) | 0 [0–2] | 7.5 [4–14] | <0.001 |
| Plasma transfusion (mL) | 0 [0–400] | 500 [0–1,000] | <0.001 |
| Platelet transfusion (U) | 0 [0–0] | 0 [0–1] | <0.001 |
| Temporary pacemaker implantation | 16 (3.8) | 11 (19.0) | <0.001 |
| Drain duration (days) | 2.2±0.7 | 4.6±4.0 | <0.001 |
| Ventilator assistance time (hours) | 20.7±14.0 | 113.0±227.5 | <0.001 |
| ICU stay (days) | 23.1±15.1 | 93.9±137.1 | <0.001 |
| Postoperative hospital stay (days) | 6.6±2.4 | 12.1±8.2 | <0.001 |
| Total hospital stay (days) | 14.9±5.5 | 23.9±13.2 | <0.001 |
Values are presented as numbers (percentages), mean ± standard deviation, or median (interquartile range). CPB, cardiopulmonary bypass; ICU, intensive care unit; MACE, major adverse cardiovascular events; RBC, red blood cell.
The overall 30-day mortality rate was 2.08% among all patients. Of these, four developed respiratory failure, and three died of septic shock caused by infection. In addition, the family of a patient opted to discontinue treatment following a stroke. Two patients died during surgery, with one experiencing circulatory failure because of iatrogenic aortic dissection and the other having treatment discontinued by their family because of resuscitation difficulties during the surgery. Detailed information on MACE occurrences in all patients is shown in Table 3.
Table 3. Individual factors of MACE.
| MACE category | Values (N=58) |
|---|---|
| Low cardiac output syndrome | 6 (10.3) |
| Bleeding complications | 19 (32.8) |
| Renal dysfunction | 4 (6.9) |
| Stroke and other cerebrovascular events | 6 (10.3) |
| Pulmonary complications | 20 (34.5) |
| Vascular complications | 3 (5.2) |
Values are presented as numbers (percentages). MACE, major adverse cardiovascular events.
Associations of clinical variable with MACE status
By using the Cox regression analysis, 31 factors independently associated with MACE were identified from the 79 clinical variables collected (Table S1). Among the baseline characteristics, older patients [Crude HR (95% CI): 1.62 (1.20–2.18), P=0.001] had a significantly higher risk of experiencing MACE. In addition, severe tricuspid valve regurgitation [Crude HR (95% CI): 1.83 (1.36–2.44), P<0.001] and increased pulmonary arterial pressure [Crude HR (95% CI): 1.66 (1.24–2.22), P=0.001] were associated with an increased risk of MACE.
In surgery, the implantation of a temporary pacemaker [Crude HR (95% CI): 4.91 (2.60–9.29), P<0.001], prolonged surgical duration [Crude HR (95% CI): 1.79 (1.56–2.05), P<0.001], intraoperative plasma transfusion [Crude HR (95% CI): 1.75 (1.58–1.94), P<0.001], prolongation of surgical time [Crude HR (95% CI): 1.65 (1.45–1.88), P<0.001], and 24 h drain output [Crude HR (95% CI): 1.72 (1.48–1.99), P<0.001] were all correlated with a higher MACE incidence.
Notably, a history of chronic liver disease [Crude HR (95% CI): 4.44 (1.91–10.34), P=0.001] combined with elevated preoperative alanine aminotransferase (AST) levels [Crude HR (95% CI): 1.26 (1.13–1.39), P<0.001] were associated with MACE occurrence. Moreover, the postoperative decrease in albumin level [Crude HR (95% CI): 0.63 (0.47–0.83), P=0.001] influenced MACE occurrence.
Model development and validation
To enable early clinical prediction of perioperative MACE in patients undergoing MIMVS, we developed the CompliMit Score using an elastic net model that screened 29 preoperative and intraoperative clinical variables related to MACE (P<0.005). The CompliMit Score comprised 14 variables: temporary pacemaker implantation, previous PCI, chronic liver disease, platelet in, red blood cell (RBC) in, surgical time, pulmonary arterial hypertension, postoperative lactate, postoperative blood urea nitrogen (BUN), preoperative AST, heart rate, CPB time, plasma in, and 24 h drain output. Figure 3 illustrates the contributions of the 14 variables selected by the elastic net model in predicting MACE. The Kaplan-Meier curve showed that using the tertiles of the CompliMit score effectively distinguished between patients with different MACE risk levels (Figure 4).
Figure 3.
Elastic Net coefficients built on clinical variables for MACE. By using the Elastic Net algorithm, 14 variables were selected from 29 clinical variables to build a model predicting MACE events. A lollipop chart displays the coefficients of these 14 variables in the model. AST, aspartate aminotransferase; BUN, blood urea nitrogen; CPB, cardiopulmonary bypass; HR, heart rate; MACE, major adverse cardiovascular events; PAH, pulmonary arterial hypertension; PCI, percutaneous coronary intervention; RBC, red blood cell.
Figure 4.
Incidence of MACE according to the CompliMit Score. Kaplan-Meier curves illustrating the timing of MACE across the three strata defined by the CompliMit Score tertiles. MACE, major adverse cardiovascular events.
The internal validation with 200 bootstrap samples of the total cohort (n=480) revealed an AUC of 0.91 (95% CI: 0.84–0.98) (Figure S1). A calibration plot of the entire population is shown in Figure S2. Furthermore, we compared the predictive power of the CompliMit Score with EuroSCORE II (Figure 5). The CompliMit Score showed better predictive accuracy [AUC: CompliMit Score, 0.92 (95% CI: 0.88–0.96); EuroSCORE II, 0.67 (95% CI: 0.59–0.75)]. In addition, the decision curve analysis further confirmed that the CompliMit Score has greater clinical utility than EuroSCORE II. For a decision threshold of 33% MACE risk, the Elastic network model would identify ~61 additional cases, without identifying any additional false positive, in a population of 1,000 patients with a 12% MACE incidence, compared with EuroSCORE II.
Figure 5.
Comparison of the CompliMit Score and EuroSCORE II in predicting MACE. (A) ROC curves for CompliMit Score and EuroSCORE II for MACE. Data are presented as AUC (95% CI). (B) Decision curve analysis of CompliMit Score vs. EuroSCORE II for MACE. AUC, area under the curve; CI, confidence interval; EuroSCORE, European System for Cardiac Operative Risk Evaluation; MACE, major adverse cardiovascular events; ROC, receiver operating characteristic.
Discussion
We gathered clinical data from 480 patients who underwent MIMVS at Anzhen Hospital in Beijing between April 2010 and May 2024. By using Cox regression analysis and machine learning techniques, we successfully identified 14 independent MACE predictors, including temporary pacemaker implantation, history of chronic liver disease, and history of PCI. Based on this, we developed a predictive model called “CompliMit Score”, which has demonstrated excellent forecasting capabilities during internal validation.
However, reoperation for cardiac surgery was not included as an independent MACE predictor. This may be attributed to the advantages of the right intercostal approach in MIMVS. This approach can reduce the risk of heart rupture, damage to major blood vessels, and aortic injury caused by a resternotomy (13). However, severe adhesions in the pleural cavity render surgical exposure extremely challenging in patients requiring multiple thoracotomies.
Our study has found that the implantation of a temporary pacemaker is a strong predictor of MACE. Temporary pacemakers are used to maintain hemodynamic stability in patients with bradyarrhythmias, such as sinus node dysfunction, sinus arrest, or atrioventricular block. These arrhythmic events may independently increase the risk of MACE, including thromboembolism and heart failure. Moreover, complications associated with temporary pacing wire placement (e.g., infection or bleeding) may indirectly contribute to MACE occurrence (14). Additionally, long-term right ventricular pacing induces left ventricular electromechanical dyssynchrony, which can trigger adverse cardiac remodeling and subsequently elevate the risk of heart failure (15).
A hybrid concept consisting of PCI and MIMVS has been reported to be beneficial in diverse patient cohorts with mitral valve and coronary disease (16-18). Based on our findings, a history of PCI was determined to be a separate risk factor for MACE, which can be attributed to two factors (19). First, data from the New York Cardiac Surgery Reporting System revealed that mortality rates ranged from 3.3% to 12.8% when patients undergoing valve surgery also had coronary artery disease (20,21). Second, patients with diseased coronary vessels left untreated experienced a higher occurrence of perioperative myocardial infarction (22). Therefore, it is essential to guarantee sufficient coronary perfusion prior to surgery or to establish appropriate myocardial protection with cardioplegia to prevent periprocedural myocardial infarction. Another possible reason is the requirement for dual antiplatelet treatment following PCI. Patients who received this treatment and underwent cardiac surgery had higher rates of postoperative bleeding, as reported from randomized controlled trials and retrospective studies (17,23-27). However, no authoritative guidelines are available to determine the optimum timeframe between PCI and MIMVS. Finding a balance between the risk of bleeding and stent thrombosis is vital for decreasing MACE occurrence.
Our analysis confirmed that abnormal liver function may increase MACE incidence in patients with MIMVS. However, the prognostic value of this risk factor has not been fully evaluated in EuroSCORE II. Specifically, hypoalbuminemia (<3.5 g/dL) has been shown to increase infection risk and delay wound healing [relative risk (RR) =1.8], likely as a result of impaired immune response and reduced tissue repair capacity (28). Some studies have suggested that serum albumin levels play a causal role in specific cardiovascular diseases (29-31). Furthermore, elevated transaminase levels (ALT/AST >1,000 U/L), a marker of acute cardiogenic hepatic ischemia, are strongly associated with 30-day mortality [odds ratio (OR) =3.1], reflecting the critical impact of hemodynamic instability on liver perfusion (32). In addition, the risk of MACE was found to differ depending on baseline serum albumin and total cholesterol levels (33). These discrepancies may be related to biases caused by the measurement techniques, confounding factors, and causal relationships.
Blood transfusion status, surgical time, and CPB time can provide a concise overview of the surgical circumstances. The methods used in surgery, including CPB management, strategies for safeguarding the heart muscle, and precision of surgical procedures, directly affect the recovery process and the potential for postoperative complications. In addition, postoperative monitoring indicators of liver, kidney, and respiratory functions are equally crucial for predicting MACE.
The 30-day mortality rates in this study are similar to those reported previously (5). Respiratory failure and septic shock secondary to pulmonary infection remain the most common causes of death, accounting for over 50% of mortality. This finding emphasizes the significance of lung protection and indicates the ongoing need for improvement in surgical techniques, patient selection, intraoperative management, and postoperative care. Lung injury can be caused or worsened by physical compression of the lungs during surgery, lung collapse during ventilation of one lung, and lung ischemia–reperfusion injury. Notably, despite the low frequency of iatrogenic aortic dissection, the mortality rate is exceptionally high, presenting a grave danger to the lives of patients (34). Performing thorough vascular screening before surgery can greatly reduce the likelihood of these issues. Patients with developmental abnormalities or large plaques in the iliac artery–femoral artery segment and those with diffuse aortic disease should carefully consider the choice of arterial catheterization method (35).
The occurrence of MACE reflects certain technical challenges and unpredictable risks, even in minimally invasive surgeries. Consequently, thorough preoperative evaluation, precision of intraoperative procedures, and close postoperative monitoring are of utmost importance.
In summary, the CompliMit Score provides a more comprehensive understanding of the technical characteristics and intraoperative management of minimally invasive surgery than the existing scoring systems. Our study not only provides a scientific basis for creating more precise and customized risk prediction tools but also sets higher standards for future therapy.
In this study, we used a single-center prognostic design, which might restrict the applicability of the findings given the specific attributes of the healthcare setting and patient demographics. Furthermore, this study did not categorize certain special conditions, such as aortitis or those who had received liver or kidney transplants, as separate risk factors. These conditions were rare within the dataset, and no related mortality was documented. In addition, the risk assessment model proposed in this study has not been externally validated in other independent patient cohorts, indicating the need for validation in different medical centers and diverse patient populations.
This study identified numerous preoperative, intraoperative, and postoperative clinical variables associated with MACE, leading to the development of a new perioperative MACE risk assessment tool, the CompliMit Score. This finding offers a framework of clinical variables that clinicians should pay special attention to during the perioperative period. This framework can be used to aid in the early identification of high-risk patients with MACE and the development of more suitable treatment strategies. Future multicenter studies are required to validate the CompliMit Score across diverse populations and surgical procedures.
Conclusions
This study identified a range of preoperative, intraoperative, and postoperative clinical variables associated with MACE, leading to the development of the CompliMit Score, a novel perioperative risk assessment tool. This score provides a structured framework for clinicians to identify high-risk patients early in the perioperative period, facilitating more tailored treatment strategies and improving patient outcomes. To ensure its robustness and applicability, future multicenter studies are required to validate the CompliMit Score across diverse populations and surgical procedures.
Supplementary
The article’s supplementary files as
Acknowledgments
We would like to thank Editage (www.editage.cn) for English language editing.
Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of Beijing Anzhen Hospital (No. KS2023092) and individual consent for this retrospective analysis was waived.
Footnotes
Reporting Checklist: The authors have completed the TRIPOD + AI reporting checklist. Available at https://cdt.amegroups.com/article/view/10.21037/cdt-2025-25/rc
Funding: None.
Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://cdt.amegroups.com/article/view/10.21037/cdt-2025-25/coif). The authors have no conflicts of interest to declare.
Data Sharing Statement
Available at https://cdt.amegroups.com/article/view/10.21037/cdt-2025-25/dss
References
- 1.Carpentier A, Loulmet D, Carpentier A, et al. Open heart operation under videosurgery and minithoracotomy. First case (mitral valvuloplasty) operated with success. C R Acad Sci III 1996;319:219-23. [PubMed] [Google Scholar]
- 2.Salman J, Franz M, Aburahma K, et al. Hypothermic Ventricular Fibrillation in Redo Minimally Invasive Mitral Valve Surgery: A Promising Solution for a Surgical Challenge. J Clin Med 2024;13:4269. 10.3390/jcm13144269 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Cote CL, Alkhamees N, Goldbach M, et al. Minimally Invasive En Bloc Resection of Massive Obstructive Atrial Tumour. Can J Cardiol 2016;32:1575.e21-3. 10.1016/j.cjca.2016.02.046 [DOI] [PubMed] [Google Scholar]
- 4.Ricci D, Boffini M, Barbero C, et al. Minimally invasive tricuspid valve surgery in patients at high risk. J Thorac Cardiovasc Surg 2014;147:996-1001. 10.1016/j.jtcvs.2013.03.018 [DOI] [PubMed] [Google Scholar]
- 5.Baccanelli F, Albano G, Carrara A, et al. Risk Factors for Higher Postoperative Myocardial Injury in Minimally Invasive Mitral Valve Surgery Patients: A Cohort Study. J Clin Med 2024;13:1591. 10.3390/jcm13061591 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Gupta S, McEwen C, Eqbal A, et al. Minimally Invasive Surgery for Congenital Heart Disease. Ann Thorac Surg 2024;118:953-62. 10.1016/j.athoracsur.2023.11.032 [DOI] [PubMed] [Google Scholar]
- 7.Koch NA, Chiappini J, Ihringer LM, et al. Long-Term Results in Minimally Invasive Non-Resectional Mitral Valve Repair for Barlow Mitral Valve Disease. J Clin Med 2025;14:1005. 10.3390/jcm14031005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Nashef SA, Roques F, Sharples LD, et al. EuroSCORE II. Eur J Cardiothorac Surg 2012;41:734-44; discussion 744-5. 10.1093/ejcts/ezs043 [DOI] [PubMed] [Google Scholar]
- 9.Van Praet KM, Kofler M, Sündermann SH, et al. Minimally invasive approach for infective mitral valve endocarditis. Ann Cardiothorac Surg 2019;8:702-4. 10.21037/acs.2019.07.01 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Guedeney P, Chevrot G, Collet JP. VARC-3 Criteria: Adding Prognosis to Injury. JACC Cardiovasc Interv 2023;16:1233-5. 10.1016/j.jcin.2023.04.012 [DOI] [PubMed] [Google Scholar]
- 11.Franzone A, Esposito G. VARC-3 Criteria for Bleeding Events: Integrating Theory Into Practice. JACC Cardiovasc Interv 2023;16:2275-6. 10.1016/j.jcin.2023.08.011 [DOI] [PubMed] [Google Scholar]
- 12.Stone GW, Vahanian AS, Adams DH, et al. Clinical Trial Design Principles and Endpoint Definitions for Transcatheter Mitral Valve Repair and Replacement: Part 1: Clinical Trial Design Principles: A Consensus Document From the Mitral Valve Academic Research Consortium. J Am Coll Cardiol 2015;66:278-307. 10.1016/j.jacc.2015.05.046 [DOI] [PubMed] [Google Scholar]
- 13.Akowuah EF, Maier RH, Hancock HC, et al. Minithoracotomy vs Conventional Sternotomy for Mitral Valve Repair: A Randomized Clinical Trial. JAMA 2023;329:1957-66. 10.1001/jama.2023.7800 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Manuel L. Temporary epicardial pacing wires post-cardiac surgery: a literature review. Gen Thorac Cardiovasc Surg 2022;70:595-601. 10.1007/s11748-022-01831-5 [DOI] [PubMed] [Google Scholar]
- 15.Bruno F, Munoz Pousa I, Saia F, et al. Impact of Right Ventricular Pacing in Patients With TAVR Undergoing Permanent Pacemaker Implantation. JACC Cardiovasc Interv 2023;16:1081-91. 10.1016/j.jcin.2023.02.003 [DOI] [PubMed] [Google Scholar]
- 16.Santana O, Xydas S, Williams RF, et al. Hybrid approach of percutaneous coronary intervention followed by minimally invasive mitral valve surgery: a 5-year single-center experience. J Thorac Dis 2017;9:S595-601. 10.21037/jtd.2017.06.29 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Santana O, Xydas S, Williams RF, et al. Outcomes of a hybrid approach of percutaneous coronary intervention followed by minimally invasive aortic valve replacement. J Thorac Dis 2017;9:S569-74. 10.21037/jtd.2017.04.28 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Leacche M, Umakanthan R, Zhao DX, et al. Surgical update: hybrid procedures, do they have a role? Circ Cardiovasc Interv 2010;3:511-8. 10.1161/CIRCINTERVENTIONS.110.957951 [DOI] [PubMed] [Google Scholar]
- 19.Moscoso-Ludueña M, Vondran M, Irqsusi M, et al. Combined Minimally Invasive Mitral Valve Surgery and Percutaneous Coronary Intervention: A Hybrid Concept for Patients with Mitral Valve and Coronary Pathologies. J Clin Med 2023;12:5553. 10.3390/jcm12175553 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Hannan EL, Racz MJ, Jones RH, et al. Predictors of mortality for patients undergoing cardiac valve replacements in New York State. Ann Thorac Surg 2000;70:1212-8. 10.1016/S0003-4975(00)01968-8 [DOI] [PubMed] [Google Scholar]
- 21.Bruno VD, Zakkar M, Guida G, et al. Combined Degenerative Mitral Valve and Coronary Surgery: Early Outcomes and 10-Year Survival. Ann Thorac Surg 2020;110:1527-33. 10.1016/j.athoracsur.2020.03.009 [DOI] [PubMed] [Google Scholar]
- 22.Leviner DB, Torregrossa G, Puskas JD. Incomplete revascularization: what the surgeon needs to know. Ann Cardiothorac Surg 2018;7:463-9. 10.21037/acs.2018.06.07 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Hensley NB, Ariyo PT, Holmes SD, et al. Center Variation in Use of Preoperative Dual Antiplatelet Therapy and Platelet Function Testing at the Time of Coronary Artery Bypass Grafting in Maryland. Anesth Analg 2022;135:567-75. 10.1213/ANE.0000000000005964 [DOI] [PubMed] [Google Scholar]
- 24.Kremke M, Gissel MS, Jensen MJ, et al. The association between a three-day ticagrelor discontinuation and perioperative bleeding complications. Eur J Cardiothorac Surg 2019;55:714-20. 10.1093/ejcts/ezy355 [DOI] [PubMed] [Google Scholar]
- 25.de Waard D, Fagan A, Minnaar C, et al. Management of patients after coronary artery bypass grafting surgery: a guide for primary care practitioners. CMAJ 2021;193:E689-94. 10.1503/cmaj.191108 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Matejic-Spasic M, Hassan K, Thielmann M, et al. Management of perioperative bleeding risk in patients on antithrombotic medications undergoing cardiac surgery-a systematic review. J Thorac Dis 2022;14:3030-44. 10.21037/jtd-22-428 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Santana O, Pineda AM, Cortes-Bergoderi M, et al. Hybrid approach of percutaneous coronary intervention followed by minimally invasive valve operations. Ann Thorac Surg 2014;97:2049-55. 10.1016/j.athoracsur.2014.02.039 [DOI] [PubMed] [Google Scholar]
- 28.Lim MH, Lee CH, Ju MH, et al. Influence of hepatic dysfunction in patients who underwent tricuspid valve surgery. J Thorac Dis 2023;15:1614-26. 10.21037/jtd-22-1741 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Huang T, An Z, Huang Z, et al. Serum albumin and cardiovascular disease: a Mendelian randomization study. BMC Cardiovasc Disord 2024;24:196. 10.1186/s12872-024-03873-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Hou XZ, Liu EQ, Liu SQ, et al. The negative association between serum albumin levels and coronary heart disease risk in adults over 45 years old: a cross-sectional survey. Sci Rep 2023;13:672. 10.1038/s41598-023-27974-w [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Nakamura N, Honzawa Y, Nishimon S, et al. Combined serum albumin, fecal immunochemical test, and leucine-rich alpha-2 glycoprotein levels for predicting prognosis in remitting patients with ulcerative colitis. Sci Rep 2023;13:13863. 10.1038/s41598-023-41137-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Ostovar R, Schroeter F, Erb M, et al. Liver cirrhosis: Still an elusive risk factor in the current EuroSCORE system. Eur J Cardiothorac Surg 2022;62:ezac128. 10.1093/ejcts/ezac128 [DOI] [PubMed] [Google Scholar]
- 33.Yao YF, Chen ZY, Luo TY, et al. Cholesterol affects the relationship between albumin and major adverse cardiac events in patients with coronary artery disease: a secondary analysis. Sci Rep 2022;12:12634. 10.1038/s41598-022-16963-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Leontyev S, Borger MA, Legare JF, et al. Iatrogenic type A aortic dissection during cardiac procedures: early and late outcome in 48 patients. Eur J Cardiothorac Surg 2012;41:641-6. 10.1093/ejcts/ezr070 [DOI] [PubMed] [Google Scholar]
- 35.Schaefer AK, Distelmaier K, Riebandt J, et al. Access site complications of postcardiotomy extracorporeal life support. J Thorac Cardiovasc Surg 2022;164:1546-1558.e8. 10.1016/j.jtcvs.2021.09.074 [DOI] [PubMed] [Google Scholar]