Epicardial placement of human placental membrane allografts in coronary artery bypass graft surgery is associated with reduced postoperative atrial fibrillation: a pilot study for a future multi-center randomized controlled trial.

BACKGROUND: Post-operative atrial fibrillation (POAF) occurs in up to 40% of patients following coronary artery bypass grafting (CABG) and is associated with a higher risk of stroke and mortality. This study investigates how POAF may be mitigated by epicardial placement of aseptically processed human placental membrane allografts (HPMAs) before pericardial closure in CABG surgery. This study was conducted as a pilot feasibility study to collect preliminary for a forthcoming multi-center randomized controlled trial. METHODS: This retrospective observational study of patients undergoing CABG surgery excluded patients with pre-operative heart failure, chronic kidney disease, or a history of atrial fibrillation. The "treatment" group (n = 24) had three HPMAs placed epicardially following cardiopulmonary bypass decannulation but before partial pericardial approximation and chest closure. The only difference in clinical protocol for the control group (n = 54) was that they did not receive HPMA. RESULTS: HPMA-treated patients saw a significant, greater than four-fold reduction in POAF incidence compared to controls (35.2-8.3%, p = 0.0136). Univariate analysis demonstrated that HPMA treatment was associated with an 83% reduction in POAF (OR = 0.17, p = 0.0248). Multivariable analysis yielded similar results (OR = 0.07, p = 0.0156) after controlling for other covariates. Overall length of stay (LOS) between groups was similar, but ICU LOS trended lower with HPMA treatment (p = 0.0677). Post-operative inotrope and vasopressor requirements were similar among groups. There was no new-onset post-operative heart failure, stroke, or death reported up to thirty days in either group. CONCLUSIONS: Epicardial HPMA placement can be a simple intervention at the end of CABG surgery that may provide a new approach to reduce post-operative atrial fibrillation by modulating local inflammation, possibly reducing ICU and hospital stay, and ultimately improving patient outcomes.

Redo partial aortic root remodelling in type A dissection with an aorto-left atrial fistula.

A 72-year-old male with a history of a triple-vessel coronary artery bypass graft years ago presented with a DeBakey type 2 aortic dissection and an aorto-left atrial fistula with patent bypass grafts (left internal mammary artery and saphenous vein grafts). He developed pulmonary oedema and required intubation. The right axillary artery was cannulated. After the ascending aorta and left internal mammary artery were clamped, the aorta was transected, leaving aortic tissue around two saphenous vein grafts as two separate patches. An entry tear was found adjacent to the proximal anastomosis of the saphenous vein graft to the posterior descending artery. A fistula, which was located between a false lumen in the non-coronary sinus and the dome of the left atrium, was primarily closed. Because the adventitia was thinned out in the non-coronary sinus due to aortic dissection, partial aortic root remodelling was performed with resuspension of the commissures. Hemiarch repair was performed under moderate hypothermia and unilateral antegrade cerebral perfusion. After systemic perfusion was resumed, the locations of the saphenous vein graft buttons were determined. The ascending graft was cross-clamped again; the saphenous vein graft to the obtuse marginal branch graft was reimplanted using the Carrel patch technique while a saphenous vein graft to the posterior descending artery required interposition of a 10-mm Dacron graft to accommodate the length.

Continuous Superficial Parasternal Intercostal Plane Catheters for Poststernotomy Pain Control: A Case Series.

This case series describes the safety and effectiveness of superficial parasternal intercostal plane catheters for poststernotomy pain control in 4 patients who underwent multivessel coronary artery bypass grafting. Patients had reduced sternal pain and opioid consumption while the catheters ran continuously for 72 hours without complications. Our experience suggests the effectiveness of parasternal blocks can be safely prolonged with catheters, and they can be a useful addition to pain management strategies for this patient population.

Surgical management of right hepatectomy after coronary artery bypass grafting using the right gastroepiploic artery: a case report and literature review.

BACKGROUND: Coronary artery bypass grafting (CABG) using the right gastroepiploic artery (RGEA) is a well-established, safe procedure. However, problems with RGEA grafts in subsequent abdominal surgeries can lead to fatal complications. This report presents the first case of right hepatectomy for hepatocellular carcinoma after CABG using the RGEA. CASE PRESENTATION: We describe a case in which a right hepatectomy for an 81-year-old male patient with hepatocellular carcinoma was safely performed after CABG using a RGEA graft. Preoperatively, three-dimensional computed tomography (3D- CT) images were constructed to confirm the run of the RGEA graft. The operation was conducted with the standby of a cardiovascular surgeon if there was a problem with the RGEA graft. The RGEA graft had formed adhesions with the hepatic falciform ligament, necessitating meticulous dissection. After the right hepatectomy, the left hepatic lobe descended into the vacated space, exerting traction on the RGEA. However, this traction was mitigated by suturing the hepatic falciform ligament to the abdominal wall, ensuring stability of the RGEA. There were no intraoperative or postoperative complications. CONCLUSION: It is crucial to confirm the functionality and anatomy of the RGEA graft preoperatively, handle it gently intraoperatively, and collaborate with cardiovascular surgeons.

Prediction of Vasoactive-Inotropic Score on Prolonged Mechanical Ventilation in Adult Congenital Heart Disease Patients After Surgical Treatment Combined with Coronary Artery Bypass Grafting.

INTRODUCTION: This study aimed to investigate the predictive value of the vasoactive-inotropic score (VIS) at different time points for postoperative prolonged mechanical ventilation (PMV) in adult congenital heart disease patients undergoing surgical treatment combined with coronary artery bypass grafting. METHODS: Patients were divided into two groups that developed PMV or not. The propensity score matching method was applied to reduce the effects of confounding factors between the two groups. VIS at different time points (VIS at the end of surgery, VIS6h, VIS12h, and VIS12h max) after surgery were recorded and calculated. The value of VIS in predicting PMV was analyzed by the receiver operating characteristic (ROC) curve, and multivariate logistic regression was used to analyze independent risk factors. RESULTS: Among 250 patients, 52 were in the PMV group, and 198 were in the non-PMV group. PMV rate was 20.8%. After propensity score matching, 94 patients were matched in pairs. At each time point, the area under the ROC curve predicted by VIS for PMV was > 0.500, among which VIS at the end of surgery was the largest (0.805). The optimal cutoff point for VIS of 6.5 could predict PMV with 78.7% sensitivity and 72.3% specificity. VIS at the end of surgery was an independent risk factor for PMV (odds ratio=1.301, 95% confidence interval 1.091~1.551, P<0.01). CONCLUSION: VIS at the end of surgery is an independent predictor for PMV in patients with adult congenital heart disease surgical treatment combined with coronary artery bypass grafting.

Comparison of the Effects of Full Median Sternotomy vs. Mini-Incision on Postoperative Pain in Cardiac Surgery: A Meta-Analysis.

INTRODUCTION: It is not yet clear whether cardiac surgery by mini-incision (minimally invasive cardiac surgery [MICS]) is overall less painful than the conventional approach by full sternotomy (FS). A meta-analysis is necessary to investigate polled results on this topic. METHODS: PubMed®/MEDLINE, Cochrane CENTRAL, Latin American and Caribbean Health Sciences Literature (or LILACS), and Scientific Electronic Library Online (or SciELO) were searched for all clinical trials, reported until 2022, comparing FS with MICS in coronary artery bypass grafting (CABG), mitral valve surgery (MVS), and aortic valve replacement (AVR), and postoperative pain outcome was analyzed. Main summary measures were the method of standardized mean differences (SMD) with a 95% confidence interval (CI) and P-values (considered statistically significant when < 0.05). RESULTS: In AVR, the general estimate of postoperative pain effect favored MICS (SMD 0.87 [95% CI 0.04 to 1.71], P=0.04). However, in the sensitivity analysis, there was no difference between the groups (SMD 0.70 [95% CI -0.69 to 2.09], P=0.32). For MVS, it was not possible to perform a meta-analysis with the included studies, because they had different methodologies. In CABG, the general estimate of the effect of postoperative pain did not favor any of the approaches (SMD -0.40 [95% CI -1.07 to 0.26], P=0.23), which was confirmed by sensitivity analysis (SMD -0.02 [95% CI -0.71 to 0.67], P=0.95). CONCLUSION: MICS was not globally less painful than the FS approach. It seems that postoperative pain is more related to the degree of tissue retraction than to the size of the incision.

[Fifteen-year results of bilateral and unilateral internal mammary artery grafting]. / Pyatnadtsatiletnie rezul'taty revaskulyarizatsii miokarda s ispol'zovaniem dvukh i odnoi vnutrennei grudnoi arterii.

OBJECTIVE: To evaluate the clinical effectiveness of bilateral internal mammary artery grafting over long-term (15 years) postoperative period. MATERIAL AND METHODS: There were 276 patients divided into two groups: 135 patients (group A) underwent bilateral internal mammary artery grafting and 141 patients (group B) underwent unilateral internal mammary artery grafting together with venous bypass grafts. On-pump surgeries and cardioplegia, parallel CPB and on-pump procedures were performed in equal proportions. Mean age of patients was 57.3±7.6 years. Diabetes mellitus was detected in 21 (15.5%) and 24 (19.1%) patients, respectively (p>0.05). Mean LV ejection fraction was 55.4±9.9%, revascularization index - 3.1±0.8 and 3.0±0.7, respectively. In the 1st group, 43 patients underwent bilateral internal mammary artery grafting alone. Autovenous grafts were additionally used in other 84 patients. RESULTS: Ten-year survival exceeded 90% in both groups. Freedom from adverse cardiac events after 15 years was significantly higher in group A (77.3% vs. 59.3%, p=0.018). In group A, 16 patients died throughout this period due to cancer (50%), myocardial infarction (12.5%), stroke (18.8%) and complications of diabetes mellitus (6.3%). In group B, 22 patients died mainly from cardiac causes (myocardial infarction - 40.9%, cancer - 27.3%). CONCLUSION: Bilateral internal mammary artery grafting has obvious advantages over traditional coronary artery bypass grafting. If we take into account higher proportion of cardiac causes in structure of mortality in group B, we can talk about positive impact of bilateral internal mammary artery grafting not only on the quality of life, but also on life expectancy in long-term postoperative period.

Quantitative flow ratio versus fractional flow reserve for Heart Team decision-making in multivessel disease: the randomised, multicentre DECISION QFR trial.

BACKGROUND: Vessel-level physiological data derived from pressure wire measurements are one of the important determinant factors in the optimal revascularisation strategy for patients with multivessel disease (MVD). However, these may result in complications and a prolonged procedure time. AIMS: The feasibility of using the quantitative flow ratio (QFR), an angiography-derived fractional flow reserve (FFR), in Heart Team discussions to determine the optimal revascularisation strategy for patients with MVD was investigated. METHODS: Two Heart Teams were randomly assigned either QFR- or FFR-based data of the included patients. They then discussed the optimal revascularisation mode (percutaneous coronary intervention [PCI] or coronary artery bypass grafting [CABG]) for each patient and made treatment recommendations. The primary endpoint of the trial was the level of agreement between the treatment recommendations of both teams as assessed using Cohen's kappa. RESULTS: The trial included 248 patients with MVD from 10 study sites. Cohen's kappa in the recommended revascularisation modes between the QFR and FFR approaches was 0.73 [95% confidence interval {CI} : 0.62-0.83]. As for the revascularisation planning, agreements in the target vessels for PCI and CABG were substantial for both revascularisation modes (Cohen's kappa=0.72 [95% CI: 0.66-0.78] and 0.72 [95% CI: 0.66-0.78], respectively). The team assigned to the QFR approach provided consistent recommended revascularisation modes even after being made aware of the FFR data (Cohen's kappa=0.95 [95% CI:0.90-1.00]). CONCLUSIONS: QFR provided feasible physiological data in Heart Team discussions to determine the optimal revascularisation strategy for MVD. The QFR and FFR approaches agreed substantially in terms of treatment recommendations.

Comparison of clinical effects of coronary artery bypass grafting between left anterior small thoracotomy approach and lower-end sternal splitting approach.

OBJECTIVE: To compare the clinical effects of coronary artery bypass grafting (CABG) between the left anterior small thoracotomy (LAST) and lower-end sternal splitting (LESS) approaches for coronary artery disease. METHODS: In total, 110 patients who underwent LAST from October 2015 to December 2020 in Tianjin Chest Hospital were selected as the observation group. Patients who underwent the LESS approach during the same period were analyzed. The propensity score was calculated by a logistic regression model, and nearest-neighbor matching was used for 1:1 matching. RESULTS: The length of hospital stay and ventilator support time were significantly shorter in the LAST than LESS group. The target vessels in the obtuse marginal branch and posterior left ventricular artery branch grafts were significantly more numerous in the LAST than LESS group, but those in the right coronary artery graft were significantly less numerous in the LAST group. CONCLUSIONS: CABG using either the LAST or LESS approach is safe and effective, especially in low-risk patients. The LAST approach can achieve complete revascularization for multivessel lesions and has the advantages of less trauma and an aesthetic outcome. However, it requires a certain learning curve to master the surgical techniques and has specific surgical indications.

Comparison between propofol and total inhalational anaesthesia on cardiovascular outcomes following on-pump cardiac surgery in higher-risk patients: a randomised controlled pilot and feasibility study.

OBJECTIVES: Myocardial revascularisation and cardiopulmonary bypass (CPB) can cause ischaemia-reperfusion injury, leading to myocardial and other end-organ damage. Volatile anaesthetics protect the myocardium in experimental studies. However, there is uncertainty about whether this translates into clinical benefits because of the coadministration of propofol and its detrimental effects, restricting myocardial protective processes. METHODS: In this single-blinded, parallel-group randomised controlled feasibility trial, higher-risk patients undergoing elective coronary artery bypass graft (CABG) surgery with an additive European System for Cardiac Operative Risk Evaluation ≥5 were randomised to receive either propofol or total inhalational anaesthesia as single agents for maintenance of anaesthesia. The primary outcome was the feasibility of recruiting and randomising 50 patients across two cardiac surgical centres, and secondary outcomes included the feasibility of collecting the planned perioperative data, clinically relevant outcomes and assessments of effective patient identification, screening and recruitment. RESULTS: All 50 patients were recruited within 11 months in two centres, allowing for a 13-month hiatus in recruitment due to the COVID-19 pandemic. Overall, 50/108 (46%) of eligible patients were recruited. One patient withdrew before surgery and one patient did not undergo surgery. All but one completed in-hospital and 30-day follow-up. CONCLUSIONS: It is feasible to recruit and randomise higher-risk patients undergoing CABG surgery to a study comparing total inhalational and propofol anaesthesia in a timely manner and with high acceptance and completion rates. TRIAL REGISTRATION NUMBER: NCT04039854.

Post-pneumonectomy coronary artery bypass graft surgery.

Postpneumonectomy changes, including ipsilateral diaphragmatic elevation and rib crowding, mediastinal shifting of the heart and reduction of respiratory functional reserve by 45%-55%, make cardiac surgery challenging in terms of access to the heart and great vessels. There is a paucity of literature on coronary artery bypass graft (CABG) surgery in patients with a history of pneumonectomy. We report a case of a male in his 70s with a history of left pneumonectomy who underwent a successful on-pump CABG surgery using the left internal mammary artery graft and great saphenous vein graft. The patient was extubated postoperatively and discharged in a stable condition. With no consensus available for optimal preoperative, intraoperative and postoperative management for these high-risk patients, our case report adds to the rare literature regarding successful on-pump CABG after pneumonectomy.

The efficacy of minimally invasive coronary artery bypass grafting (mics cabg) for patients with coronary artery diseases and diabetes: a single center retrospective study.

BACKGROUND: conventional coronary artery bypass grafting (CCABG) tends to cause severe complications in patients with comorbid Coronary Artery Diseases (CAD) and diabetes. On the other hand, the Minimally Invasive Cardiac Surgery Coronary Artery Bypass Grafting (MICS CABG) via transthoracic incision is associated with rapid recovery and reduced complications. Adding to the limited literature, this study compares CCABG and MICS CABG in terms of efficacy and safety. METHODS: Herein, 104 CCABG and MICS CABG cases (52 cases each) were included. The patients were recruited from the Minimally Invasive Cardiac Surgery Center, Anzhen Hospital, between January 2017 and December 2021 and were selected based on the Propensity Score Matching (PSM) model. The key outcomes included All-cause Death, Myocardial Infarction (MI), Cerebrovascular Events, revascularization, Adverse Wound Healing Events and one-year patency of the graft by coronary CTA. RESULTS: Compared to CCABG, MICS CABG had longer surgical durations [4.25 (1.50) h vs.4.00 (1.13) h, P = 0.028], but showed a reduced intraoperative blood loss [600.00 (400.00) mL vs.700.00 (300.00) mL, P  = 0.032] and a lower secondary incision debridement and suturing rate (5.8% vs.19.2%, P = 0.038). In follow up, no statistically significant differences were found between the two groups in the cumulative Major Adverse Cardiovascular and Cerebrovascular Events (MACCEs) incidence (7.7% vs. 5.9%), all-cause mortality (0 vs. 0), MI incidence (1.9% vs. 2.0%), cerebral apoplexy incidence (5.8% vs. 3.9%), and repeated revascularization incidence (0 vs. 0) (P > 0.05). Additionally, coronary CTA results revealed that the two groups' one-year graft patency (94.2% vs. 90.2%, P = 0.761) showed no statistically significant difference. CONCLUSION: In patients with comorbid CAD and diabetes, MICS CABG and CCABG had comparable revascularization performances. Moreover, MICS CABG can effectively reduce, if not prevent, poor clinical outcomes/complications, including incision healing, sternal infection and prolonged length of stay in diabetes patients.

Immediate or staged complete revascularisation in patients presenting with acute coronary syndrome by number of diseased vessels: a substudy of the BIOVASC randomised trial.

BACKGROUND: In patients presenting with acute coronary syndrome (ACS), the number of diseased vessels may affect the efficacy of a complete revascularisation strategy. AIMS: The authors sought to evaluate the safety and efficacy of immediate complete revascularisation (ICR) and staged complete revascularisation (SCR) in patients presenting with ACS stratified by the number of diseased vessels. METHODS: In this prespecified analysis of the BIOVASC trial, ICR was compared with SCR in patients with two-vessel disease (2VD) or three-vessel disease (3VD). The primary endpoint was a composite of all-cause mortality, myocardial infarction (MI), any unplanned ischaemia-driven revascularisation or cerebrovascular events at 1 year after the index procedure. Comparisons were performed using Cox regression. RESULTS: A total of 1,525 patients were enrolled in the BIOVASC trial, of whom 1,177 presented with 2VD and 265 with 3VD. In the 2VD group, 613 patients were assigned to ICR and 564 to SCR. In the 3VD group, 117 patients were assigned to ICR and 148 to SCR. ICR and SCR led to similar results in both the 2VD (hazard ratio [HR] 0.76, 95% confidence interval [CI]: 0.50-1.13; p=0.18) and 3VD groups (HR 0.79, 95% CI: 0.39-1.59; p=0.51) (pinteraction=0.91) in terms of the primary endpoint. ICR was associated with a lower rate of MI in patients with 3VD (HR 0.21, 95% CI: 0.046-0.93; p=0.04) (pinteraction=0.30). CONCLUSIONS: ICR might be an option in patients presenting with extensive 3VD and might be associated with a lower rate of myocardial infarction compared with SCR.

Early Outcome of On-pump Versus Off-pump Elective Surgical Revascularization in Patients with Prior ST-Segment Elevation Myocardial Infarction.

Coronary artery bypass graft surgery (CABG) is a proven treatment for coronary artery disease. History of a ST-elevation myocardial infarction (STEMI) is considered an independent risk factor for CABG irrespective of timing for an emergency or elective surgery. Patients with STEMI are candidates for both On-pump and Off-pump CABG procedures. This paper discusses the possible best option for elective surgical revascularization in patients with prior STEMI. This prospective clinical trial of 60 eligible patients with prior STEMI was conducted in a Tertiary Care Hospital from April 2018 to March 2019. Among them, 30 patients underwent off-pump (Group A) and 30 patients underwent on-pump (Group B) CABG procedures. Outcomes between both groups were observed from surgery to 1 month postoperatively. Data was analysed by the software statistical program for social science (SPSS 25.0 Inc). The surgery was successful in both groups of patients. Differences were observed by mean number of grafts per patient (2.77±0.43 vs. 3.10±0.71) and duration of operation (4.41±0.35 hours vs. 5.71±0.48 hours). An improvement in Left Ventricular Ejection Fraction (LVEF %) was observed in both groups postoperatively (17.98% vs. 10.98%) and the postoperative LVEF% at different time points were found statistically significant (p<0.05) over preoperative LVEF%. Multivariable stepwise logistic regression analysis correlated on-pump CABG with prolonged need for ionotropic support, need for blood transfusion, longer hospital stay and less improvement in LVEF%. The study supports the Off-pump CABG as a better surgical option over on-pump CABG in patients with prior STEMI.

Type A aortic dissection after Coronary Artery Bypass Grafting.

BACKGROUND Aortic dissection is rare after coronary artery bypass grafting (CABG). It is a potentially fatal complication of cardiac surgery. Reoperation may pose problems with thoracotomy, adhesiolysis, and myocardial protection. No standard treatment guidelines exist for chronic aortic dissection after CABG. We present a case of chronic type A aortic dissection after cardiac surgery, which was successfully treated. CASE REPORT A 65-year-old female patient presented with a medical background of hypertension, type 2 diabetes mellitus, and hyperlipidemia.No connective tissue disorders were diagnosed. The aortic valve was tricuspid. Three years ago, she had undergone coronary artery bypass grafting involving four branches at a different medical facility. She was complaining of chest pain weeks after bypass surgery, which gradually increased. Aortic dissection was observed in the latest contrast-enhanced CT scan. Beginning from just above the sinotubular junction and reaching below the brachiocephalic trunk.  Two grafts from the saphenous vein were patent, and one was lying just below the sternum. Cardiopulmonary support was initiated by cannulating the right femoral artery and vein. An opening in the ascending aorta exposed an intimal tear near the proximal anastomosis of the two great saphenous vein grafts. Antegrad Del-Nido cardioplegia was given through native ostia and functional bypass grafts. Proximal and then distal anastomosis of graft prosthesis was performed. A new venous graft was anastomosed to the apical part of the left anterior descending artery (LAD). This saphenous vein and the buttons of the two previous vein grafts were anastomosed to the prosthesis. The patient was successfully liberated from the heart-lung machine and exhibited favorable cardiac function in the postoperative period. CONCLUSIONS We can conclude that initial peripheral cannulation with a half dose of heparin provides a relatively bloodless and secure re-entry sternotomy. Del-Nido cardioplegia is easy to implement, safe, and gives surgeons enough time without interruptions to perform complex procedures fluently.  .

Posterior pericardiotomy and the prevention of post-operative atrial fibrillation and cardiac tamponade in isolated coronary artery bypass grafting - A retrospective analysis.

BACKGROUND: Post-Operative Atrial Fibrillation (POAF) is the most frequent complication of cardiac surgery and is associated with reduced survival, increased rates of cognitive changes and cerebrovascular accidents, heart failure, renal dysfunction, infection, length of stay and hospital costs. Cardiac tamponade although less common, carries high morbidity and mortality. Shed mediastinal blood in the pericardial space is a major source of intrapericardial oxidative stress and inflammation that triggers POAF. The utilisation of a posterior pericardiotomy (PP) aims to shunt blood from pericardium into the pleural space and have a role in the prevention of POAF as well as cardiac tamponade. METHODS: 2168 patients had undergone isolated Coronary Artery Bypass Grafting at Royal Hobart Hospital from 2008 to 2022. They were divided into PP group vs. control group. Patient baseline demographics, intraoperative data and post-operative outcomes were reviewed retrospectively. RESULTS: Total incidence of new POAF and cardiac tamponade was 24% and 0.74% respectively. Primary outcome of both the incidence of POAF (20.2% vs. 26.3%, p < 0.05) and Cardiac Tamponade (0% vs. 1.1%, p < 0.05) were less in the pericardiotomy group. A subgroup analysis of patients with recent myocardial infarction showed reduced incidence of POAF in the PP group (p < 0.05). Increasing age, Body Mass Index, poor left ventricular ejection fraction (EF < 30%) and return to theatre were independent predictors of developing POAF. There were similar rates of return to theatre for bleeding however, no cases of tamponade in the pericardiotomy group. There were no complications attributable to left posterior pericardiotomy and the time added to the duration of surgery was minimal. CONCLUSION: Posterior pericardiotomy is associated with a significant reduction in the incidence of POAF and cardiac tamponade which is safe and efficient.

Prediction of coronary artery bypass graft outcomes using a single surgical note: An artificial intelligence-based prediction model study.

BACKGROUND: Healthcare providers currently calculate risk of the composite outcome of morbidity or mortality associated with a coronary artery bypass grafting (CABG) surgery through manual input of variables into a logistic regression-based risk calculator. This study indicates that automated artificial intelligence (AI)-based techniques can instead calculate risk. Specifically, we present novel numerical embedding techniques that enable NLP (natural language processing) models to achieve higher performance than the risk calculator using a single preoperative surgical note. METHODS: The most recent preoperative surgical consult notes of 1,738 patients who received an isolated CABG from July 1, 2014 to November 1, 2022 at a single institution were analyzed. The primary outcome was the Society of Thoracic Surgeons defined composite outcome of morbidity or mortality (MM). We tested three numerical-embedding techniques on the widely used TextCNN classification model: 1a) Basic embedding, treat numbers as word tokens; 1b) Basic embedding with a dataloader that Replaces out-of-context (ROOC) numbers with a tag, where context is defined as within a number of tokens of specified keywords; 2) ScaleNum, an embedding technique that scales in-context numbers via a learned sigmoid-linear-log function; and 3) AttnToNum, a ScaleNum-derivative that updates the ScaleNum embeddings via multi-headed attention applied to local context. Predictive performance was measured via area under the receiver operating characteristic curve (AUC) on holdout sets from 10 random-split experiments. For eXplainable-AI (X-AI), we calculate SHapley Additive exPlanation (SHAP) values at an ngram resolution (SHAP-N). While the analyses focus on TextCNN, we execute an analogous performance pipeline with a long short-term memory (LSTM) model to test if the numerical embedding advantage is robust to model architecture. RESULTS: A total of 567 (32.6%) patients had MM following CABG. The embedding performances are as follows with the TextCNN architecture: 1a) Basic, mean AUC 0.788 [95% CI (confidence interval): 0.768-0.809]; 1b) ROOC, 0.801 [CI: 0.788-0.815]; 2) ScaleNum, 0.808 [CI: 0.785-0.821]; and 3) AttnToNum, 0.821 [CI: 0.806-0.834]. The LSTM architecture produced a similar trend. Permutation tests indicate that AttnToNum outperforms the other embedding techniques, though not statistically significant verse ScaleNum (p-value of .07). SHAP-N analyses indicate that the model learns to associate low blood urine nitrate (BUN) and creatinine values with survival. A correlation analysis of the attention-updated numerical embeddings indicates that AttnToNum learns to incorporate both number magnitude and local context to derive semantic similarities. CONCLUSION: This research presents both quantitative and clinical novel contributions. Quantitatively, we contribute two new embedding techniques: AttnToNum and ScaleNum. Both can embed strictly positive and bounded numerical values, and both surpass basic embeddings in predictive performance. The results suggest AttnToNum outperforms ScaleNum. With regards to clinical research, we show that AI methods can predict outcomes after CABG using a single preoperative note at a performance that matches or surpasses the current risk calculator. These findings reveal the potential role of NLP in automated registry reporting and quality improvement.