CT angiography-based radiomics as a tool for carotid plaque characterization: a pilot study.

PURPOSES: Radiomics is a quantitative method able to analyze a high-throughput extraction of minable imaging features. Herein, we aim to develop a CT angiography-based radiomics analysis and machine learning model for carotid plaques to discriminate vulnerable from no vulnerable plaques. MATERIALS AND METHODS: Thirty consecutive patients with carotid atherosclerosis were enrolled in this pilot study. At surgery, a binary classification of plaques was adopted ("hard" vs "soft"). Feature extraction was performed using the R software package Moddicom. Pairwise feature interdependencies were evaluated using the Spearman rank correlation coefficient. A univariate analysis was performed to assess the association between each feature and the plaque classification and chose top-ranked features. The feature predictive value was investigated using binary logistic regression. A stepwise backward elimination procedure was performed to minimize the Akaike information criterion (AIC). The final significant features were used to build the models for binary classification of carotid plaques, including logistic regression (LR), support vector machine (SVM), and classification and regression tree analysis (CART). All models were cross-validated using fivefold cross validation. Class-specific accuracy, precision, recall and F-measure evaluation metrics were used to quantify classifier output quality. RESULTS: A total of 230 radiomics features were extracted from each plaque. Pairwise Spearman correlation between features reported a high level of correlations, with more than 80% correlating with at least one other feature at |ρ|> 0.8. After a stepwise backward elimination procedure, the entropy and volume features were found to be the most significantly associated with the two plaque groups (p < 0.001), with AUCs of 0.92 and 0.96, respectively. The best performance was registered by the SVM classifier with the RBF kernel, with accuracy, precision, recall and F-score equal to 86.7, 92.9, 81.3 and 86.7%, respectively. The CART classification tree model for the entropy and volume features model achieved 86.7% well-classified plaques and an AUC of 0.987. CONCLUSION: This pilot study highlighted the potential of CTA-based radiomics and machine learning to discriminate plaque composition. This new approach has the potential to provide a reliable method to improve risk stratification in patients with carotid atherosclerosis.

Isolation of clinically relevant concentrations of bone marrow mesenchymal stem cells without centrifugation.

BACKGROUND: This study examined the quality of bone marrow aspirates extracted using a novel, FDA cleared method to optimally target cells from the inner cortical iliac bone surface without the need for centrifugation. This method employs small draws from a single puncture that promote only lateral flow from multiple sites (SSLM method). The study utilized the Marrow Cellutions bone marrow aspiration system (MC system) which is based on the SSLM method and compared the MC system directly to bone marrow concentrates (BMAC) generated by centrifugation of aspirates harvested with a standard aspiration needle. METHODS: Three direct comparisons were conducted evaluating the SSLM draws and BMACs derived from the same patient from contralateral iliac crests. The levels of TNCs/mL, CD34+ cells/mL, CD117+ cells/mL, and CFU-f/mL were compared between the various bone marrow preparations. The cellular content of a series of SSLM draws was also analyzed to determine the total nucleated cell (TNC) count and the concentration of mesenchymal stem/progenitor cells as measured by colony forming unit fibroblasts (CFU-f). RESULTS: In direct comparisons with BMAC systems, SSLM draws yielded significantly higher CFU-f concentrations and comparable concentrations of CD34+ and CD117+ cells. In addition, the average quantity of TNCs/mL in a series of 30 patients utilizing the SSLM draw was 35.2 × 106 ± 17.1 × 106 and the average number of CFU-f/mL was 2885 ± 1716. There were small but significant correlations between the TNCs/mL and the CFU-fs/mL using the SSLM method as well as between the age of the patient and the CFU-fs/mL. CONCLUSIONS: The MC Device, using the SSLM draw technique, produced concentrations of CFU-fs, CD34+ cells and CD117+ cells that were comparable or greater to BMACs derived from the same patient. Given the rapid speed and simplicity of the MC Device, we believe this novel system possesses significant practical advantages to other currently available centrifugation based systems.

Full Orifice Patching without Annuloplasty for Severe Functional Tricuspid Valve Regurgitation.

Surgical treatment of severe functional tricuspid regurgitation associated with dilated right ventricle and increased chordal tethering (>8 mm) is challenging. We designed a technique where the anterior and posterior leaflets are detached from 50% of the annulus and a patch as large as the tricuspid orifice is sewn to augment the leaflets' tissue to force the coaptation with the septal leaflet. Annuloplasty is not performed, as it can only increase the chordal tethering, reducing the benefit of tissue augmentation. Early and midterm results in a subgroup of patients with unfavorable anatomical aspects are encouraging.

Endovascular treatment of persistent sciatic artery aneurysm with the multilayer stent.

PURPOSE: To describe the successful management of a symptomatic persistent sciatic artery (PSA) aneurysm using the Multilayer Aneurysm Repair System (MARS). CASE REPORT: A 50-year-old man with history of smoking, hypertension, and hypercholesterolemia suddenly developed pain in the buttock radiating to the left lower limb, mimicking sciatica. The ∼57×54-mm incomplete type PSA aneurysm was treated with two 14×80- and 14×60-mm MARS devices through a surgical left axillary artery access. There were no periprocedural complications. Thirty-day imaging documented patency of the stents and a minor reduction (∼47×55 mm) in the aneurysm, with slightly hyperdense content and no signs of perfusion. At 6 months, duplex and computed tomographic angiography showed complete thrombosis of the sac, patency of the multilayer stents, and further aneurysm shrinkage (40×37 mm). CONCLUSION: The multilayer stent can alter the hemodynamics inside a peripheral aneurysm to promote sac thrombosis and redirect flow to collaterals originating from the sac.

Autologous Bone Marrow Stem Cells in Patients With Critical Limb Ischaemia not Eligible for Revascularization: A Single Centre Experience.

We evaluated the use of autologus bone marrow stem cells transplantation in patients with critical limb ischaemia (CLI) not eligible for revascularization. Eighty consecutive patients candidate to amputation were enrolled in a single-centre retrospective study. The primary endpoint was defined as the amputation-free rate from stem cells transplantation. Secondary endpoints were the evaluation of transcutaneous oximetry and its predictive potential for probability of amputation and the evaluation of rest pain. Ankle brachial index, transcutaneous oxygen (TcpO2) and radiological imaging were performed at the enrolment and during the follow-up times. All patients were treated with auto transplant of bone marrow stem cells. Two patients died due to acute renal and acute respiratory failures. 19 patients were amputated from the thigh or leg. In total, 59 of 80 patients intended to thigh amputation saved the limb, preserving the plantar support. TcpO2 was found a predictive metric with an AUC equal to .763, and a threshold for a risk of amputation of 10% and 5% at the values ≤22.7 and ≤26.9 mmHg, respectively. Auto transplant of bone marrow stem cells seems to be a safe and an efficient option for CLI not eligible to revascularizzation.

Different methods of bone marrow harvesting influence cell characteristics and purity, affecting clinical outcomes.

Background: Bone marrow (BM)-derived stem cells were implanted to induce angiogenesis in patients with no-option critical limb-threatening ischemia. Considering the potential for this therapy, conflicting results related to BM harvesting methods have been reported that could affect stem cell concentrations and quality. Methods: A total of 75 patients with no-option critical limb-threatening ischemia were treated with BM implantation. For 58 patients, BM was harvested using a BM aspirate concentrate system (Harvest Technologies; group HT) with a standard aspiration needle, followed by an automated centrifugation process, to produce BM aspirate concentrate. For 17 patients, BM was harvested using the Marrow Cellution system (Aspire Medical Innovation; group MC). CD34+ cells/mL, CD117+ cells/mL, CD133+ cells/mL, CD309+ cells/mL, hematocrit, and BM purity were compared between the two BM preparations. Results: The retrospective analysis of a subset group after adjustment for age shows that the quality of BM obtained using the Marrow Cellution system is better, in terms of purity, than the classic harvesting method before centrifugation. Harvested BM before centrifugation is characterized by a higher percentage of CD133+ cells compared with BM after centrifugation. In contrast, the MC aspirate had a larger amount of very small embryonic-like cells, as indicated by the higher percentage of CD133+, CD34+, and CD45- cells. These differences translated into an increased occurrence of leg amputations in group HT than in group MC and an increase in transcutaneous oxygen pressure in patients treated with BM aspirated using MC. Conclusions: BM manipulation, such as centrifugation, affects the quality and number of stem cells, with detrimental consequences on clinical outcomes, as reflected by the different amputation rates between the two groups.

Combined Treatment With Carotid Endoarterectomy and Coronary Artery Bypass Grafting: A Single-Institutional Experience in 222 Patients.

INTRODUCTION: Carotid atherosclerotic disease is a known independent risk factor of post operative stroke after coronary artery bypass grafting (CABG). The best management of concomitant coronary artery disease and carotid artery disease remains debated. Current strategies include simultaneous carotid endoarterectomy (CEA) and CABG, staged CEA followed by CABG, staged CABG followed by CEA, staged transfemoral carotid artery stenting (TF-CAS) followed by CABG, simultaneous TF-CAS and CABG and transcarotid artery stenting. METHODS: We report our experience based on a cohort of 222 patients undergoing combined CEA and CABG surgery who come to our observation from 2004 to 2020. All patients with >70% carotid stenosis and severe multivessel or common truncal coronary artery disease underwent combined CEA and CABG surgery at our instituion. 30% of patients had previously remote neurological symptoms or a cerebral CT-scan with ischemic lesions. Patients with carotid stenosis >70%, either asymptomatic or symptomatic, underwent CT-scan without contrast media to assess ischemic brain injury, and in some cases, if necessary, CT-angiography of the neck and intracranial vessels. RESULTS: The overall perioperative mortality rate was 4.1% (9/222 patients). Two patients (.9%) had periprocedural ipsilateral transient ischemic attack (TIA) which completely resolved by the second postoperative day. Two patients (.9%) had an ipsilateral stroke, while 7 patients (3.2%) had a stroke of the controlateral brain hemisphere. Two patients (.9%) patients were affected by periprocedural coma caused by cerebral hypoperfusion due to perioperative heart failure. There were no statistically significant differences between patients in Extracorporeal Circulation (ECC) and Off-pump patients in the onset of perioperative stroke. CONCLUSION: Our experience reported that combined surgical treatment of CEA and CABG, possibly Off-Pump, is a feasible treatment procedure, able to minimize the risk of post-operative stroke and cognitive deficits.

Clinical and Hemodynamic Outcomes of Rapid-Deployment Aortic Bioprostheses.

Aim of this retrospective, multicenter study was to evaluate early and mid-term clinical and hemodynamic results of patients who underwent surgical aortic valve replacement (SAVR) with Intuity rapid-deployment bioprostheses (RDB) (Edwards Lifesciences, Irvine, CA). We analyzed data from the Italian Registry of Intuity Valve (INTU-ITA registry) that is a national, real-world and independent from the industry registry. Preoperative variables were defined according to EuroSCORE and postoperative outcomes according to Valve Academic Research Consortium (VARC). Survival distribution was evaluated using the Kaplan-Meier approach. A Cox-Proportional Hazard Model was employed to assess the effect of the covariates on patients' survival. The registry included 1687 patients from 23 centers (June 2012-September 2019). Aortic cross clamp time for isolated SAVR was 55 minutes (IQR: 45-70 minute). Postoperative pace-maker rate was 6.3%. At discharge transaortic peak and mean gradients were: 18 mm Hg (IQR: 14-23 mm Hg) and 10 mmHg (IQR: 8-13 mm Hg), respectively. Indexed effective orifice area was 1.10 cm2/m2 (IQR: 0.91-1.31 cm2/m2) and the incidence of severe patient-prosthesis mismatch was 0.6%. Hemodynamic data for all valve sizes remained stable during follow-up. Thirty-day overall mortality was 1.8% (30 patients), and at follow-up it was 5.3% (89 patients). Kaplan-Meier overall survival was 95.5% (94.3-96.7%); 90.7% (88.3-93.1%); 86.4% (82.6-90.4%) at 1, 3, and 5 years, respectively. Serum creatinine (HR: 1.36; 95%CI: 1.04-1.81; p = 0.0397) and cross-clamp time (HR: 1.01; 95%CI: 1.002-1.017; p = 0.0077) were identified as independent predictors of mortality. According to our data from the INTU-ITA registry, SAVR with RDB provides good early clinical and hemodynamic results that are confirmed at follow-up.

Intramural haematoma of the thoracic aorta: A case series.

We reported four cases of intramural haematoma of the descending thoracic aorta. Four patients, aged 55-82 years, hypertensive, were transferred from the emergency room of other hospitals due to the appearance of epigastric pain and left thorax pain. All patients underwent computed tomography angiography reporting the presence of intramural haematoma. Three patients underwent a drug therapy to maintain a controlled hypotension. A computed tomography revaluation was performed documenting (1) an increase in the thickness of the intramural haematoma, (2) the appearance of a penetrating ulcer within the haematoma and (3) the appearance of several penetrating lesions throughout the thoracic aorta. Patients required the placement of one or two thoracic aorta endoprosthesis. For the fourth patient, the hyperdense appearance of the intramural haematoma and the presence of pleural effusion suggested an urgent treatment intervention. All patients underwent a placement of cerebrospinal fluid catheter and drainage before treatment. All patients were treated with endovascular intervention with 100% technical success and absence of migration or retrograde type A dissection. There were no complications related to femoral surgical access or access routes. Perioperative mortality was null; no patient had paraplegia. No strokes, transient ischemic attack or perioperative myocardial infarction were observed. The average hospitalization was 5 days. After 3 months, angio-computed tomography reported for all patients a complete reabsorption of the intramural haematoma and a complete exclusion of the penetrating ulcer of the aortic wall present at the time of the intervention. There have been no cases of distant thoracic aortic tears. Endovascular treatment must be considered the preeminent treatment for thoracic aortic haematoma. Best timing to perform the endovascular procedure depends on the patient clinical picture and on stability of hemodynamic parameters.

Minimally Invasive vs Conventional Aortic Valve Replacement With Rapid-Deployment Bioprostheses.

BACKGROUND: The aim of this multicenter retrospective study was to compare early and midterm clinical and hemodynamic results of aortic valve replacement with rapid-deployment bioprostheses performed through conventional full-sternotomy vs mini-sternotomy. METHODS: Data from the Italian multicenter registry of aortic valve replacement with rapid-deployment bioprostheses (INTU-ITA registry) were analyzed. Patients were divided into 2 groups: full sternotomy (FS) and ministernotomy (MS). Primary endpoint was the comparison of early and midterm mortality. Secondary endpoints were: comparison of intraoperative variables, complications, and hemodynamic performance. A propensity score weighting approach was used for data analysis. RESULTS: A total of 1057 patients were analyzed: 435 (41.2%) and 622 (58.8%) in group FS and MS, respectively. Thirty-day mortality was 1.6% and 0.6% in FS and MS groups, respectively (P = .074). cardiopulmonary bypass time was 78.5 minutes and 83 minutes in FS and MS groups, respectively (P = .414). In the overall cohort, the incidence of intraoperative complications and of device success was 3.8% (40 patients) and 95.9% (1014 patients), respectively, with no significant differences between groups. Survival at 1, 3, and 5 years was 94.1%, 98.1%, 88.5% and 91.8%, 85.2%, and 84.8% in FS and MS groups, respectively (P = .412). The 2 groups showed similar postoperative gradients (median mean gradient, FS: 10.0 mm Hg, MS: 11.0 mm Hg; P = .170) and also similar incidence of patient-prosthesis mismatch (FS: 7%, MS: 6.4%, P = .647). CONCLUSIONS: According to our data, rapid-deployment bioprostheses allow the performance of minimally invasive aortic valve replacement with similar surgical times and similar clinical and hemodynamic outcomes to conventional surgery and should be considered the first choice in these procedures.

Cardiac surgery practice during the COVID-19 outbreak: a multicentre national survey.

OBJECTIVES: Healthcare systems worldwide have been overburdened by the coronavirus disease 2019 (COVID-19) outbreak. Accordingly, hospitals had to implement strategies to profoundly reshape both non-COVID-19 medical care and surgical activities. Knowledge about the impact of the COVID-19 pandemic on cardiac surgery practice is pivotal. The goal of the present study was to describe the changes in cardiac surgery practices during the health emergency at the national level. METHODS: A 26-question web-enabled survey including all adult cardiac surgery units in Italy was conducted to assess how their clinical practice changed during the national lockdown. Data were compared to data from the corresponding period in 2019. RESULTS: All but 2 centres (94.9%) adopted specific protocols to screen patients and personnel. A significant reduction in the number of dedicated cardiac intensive care unit beds (-35.4%) and operating rooms (-29.2%), along with healthcare personnel reallocation to COVID departments (nurses -15.4%, anaesthesiologists -7.7%), was noted. Overall adult cardiac surgery volumes were dramatically reduced (1734 procedures vs 3447; P < 0.001), with a significant drop in elective procedures [580 (33.4%) vs 2420 (70.2%)]. CONCLUSIONS: This national survey found major changes in cardiac surgery practice as a response to the COVID-19 pandemic. This experience should lead to the development of permanent systems-based plans to face possible future pandemics. These data may effectively help policy decision-making in prioritizing healthcare resource reallocation during the ongoing pandemic and once the healthcare emergency is over.

Surgical aortic valve replacement with new-generation bioprostheses: Sutureless versus rapid-deployment.

OBJECTIVES: The aim of this retrospective multicenter study was to compare early clinical and hemodynamic outcomes of Perceval-S sutureless (Livanova, London, United Kingdom) and Intuity rapid-deployment (Edwards Lifesciences, Irvine, Calif) bioprostheses. METHODS: Data from patients who underwent isolated or combined aortic valve replacement with Perceval-S and with Intuity bioprostheses at 18 cardiac surgical institutions were analyzed. Propensity matching was performed to identify similar patient cohorts. RESULTS: We included 911 patients from March 2011 until May 2017. Perceval-S and Intuity valves were implanted in 349 (38.3%) and in 562 (61.7%) patients, respectively. Propensity score identified 117 matched pairs. In the matched cohort, device success was 99.1% and 100% in Perceval-S and Intuity group, respectively (P = 1.000). Thirty-day Valve Academic Research Consortium mortality occurred in 2 (1.7%) and 4 (3.4%) patients in the Perceval-S and in Intuity group, respectively (P = .6834). The rate of postoperative new permanent pacemaker implantation was 6% (7 patients) and 6.8% (8 patients) in the Perceval-S and in Intuity group, respectively (P = .7896). Perceval-S valve implantation requires significantly shorter aortic crossclamp and cardiopulmonary bypass times than Intuity valve implantation (aortic crossclamp time for isolated, 52 ± 14 minutes vs 62 ± 24 minutes; P < .0001). Peak transaortic gradients were 22.4 ± 8.1 mm Hg and 19.6 ± 6.7 mm Hg (P = .0144), whereas mean gradients were 11.8 ± 4.7 mm Hg and 10.5 ± 3.9 mm Hg (P = .0388) in the Perceval-S and Intuity groups, respectively. CONCLUSIONS: Sutureless Perceval-S and rapid-deployment Intuity bioprostheses provide good and similar early clinical and hemodynamic outcomes. Perceval-S valve implantation requires shorter crossclamp and cardiopulmonary bypass times, whereas Intuity valve implantation provides lower transaortic peak and mean gradients.

Spontaneous rectus sheath hematoma: The utility of CT angiography.

We described the utility of computed tomography (CT) angiography in detection of bleeding vessels for a rapid percutaneous arterial embolization of the spontaneous rectus sheath hematoma. A 70-year-old woman comes to our attention with acute abdominal pain and a low hemoglobin level. An unenhanced CT was performed demonstrating a large rectus sheath hematoma. A conservative management was initially established. Despite this therapy, the abdominal pain increased together with a further decrease of hemoglobin values. A CT angiography was then performed, demonstrating an active bleeding within the hematoma and addressing the patient to a rapid percutaneous arterial embolization.

Early and Mid-Term Results of Rapid Deployment Valves: The Intuity Italian Registry (INTU-ITA).

BACKGROUND: Rapid deployment bioprostheses (RDBs) have been recently introduced into clinical practice for the treatment of severe aortic valve stenosis. The aim of this retrospective multicenter study was to assess early and mid-term clinical and hemodynamic outcomes of patients undergoing RDB implantation. METHODS: Data from a national registry that included patients who underwent isolated or combined aortic valve replacement with RDB in Italy were analyzed. Definitions of the European System for Cardiac Operative Risk Evaluation were used for preoperative variables and updated definitions from the Valve Academic Research Consortium were used for postoperative outcomes assessment. Univariable and multivariable analyses were performed to identify independent predictors of mortality. Follow-up was performed with clinical and echocardiographic examinations at each study site and, if this was not possible, through telephonic interviews. The Kaplan-Meier method was used for survival analysis. RESULTS: A total of 902 patients (December 2012 through November 2017) from 20 national centers were included in the registry. Device success was 95.9%, and 30-day all-cause mortality was 2.8%. Postoperative pacemaker implantation was needed in 63 patients (6.9%). At discharge, peak and mean transaortic gradients were 19 ± 7 mm Hg and 11 ± 4 mm Hg, respectively. Mild and moderate aortic regurgitation were found in 71 patients (8.2%) and in 10 patients (1.2%), respectively. Median follow-up time was 357 days (interquartile range: 103 to 638 days). Survival at 4 years was 86% ± 1%. Preoperative conduction disturbances and history of previous myocardial infarction were independently associated with mortality. CONCLUSIONS: Aortic RDBs provide good early and mid-term clinical and hemodynamic outcomes. These devices may be considered as a reasonable alternative to conventional bioprostheses, especially in minimally invasive and combined operations.

Pseudoaneurysm after patch-free carotid bifurcation endarterectomy: a case report.

The purpose of this report is to present the case of a 59-year-old man affected by multivessel coronary artery disease and sterile pseudoaneurysm of the right carotid bifurcation presenting as a pulsating neck mass 1 month after patch-free carotid endarterectomy. The surgical approach included median sternotomy and incision parallel to the anterior margin of the right sternocleidomastoid muscle. The pseudoaneurysm was excised after control of the brachiocefalic trunk and insertion of a Pruit-Inahara shunt, and the carotid vessels were directly sutured without using any prosthetic or autologous material. Concomitant coronary bypass grafting was performed. The postoperative course was uneventful and there was no evidence of recurrence of pseudoaneurysm at 6-month follow-up. On the basis of this experience and of pertinent literature, the options for the management of such rare entities are discussed.

Thoracic radiation therapy and suitability of internal thoracic arteries for myocardial revascularization.

INTRODUCTION AND BACKGROUND: Myocardial revascularization using internal thoracic arteries (ITAs) has been associated with superior clinical outcome. This study addresses the question of whether internal mammary arteries are unsuitable for grafting due to radiation-based damage in patients with history of thoracic radiation therapy. We review our experience in this subset of surgical candidates. PATIENTS AND METHODS: Forty-nine patients undergoing elective coronary artery bypass grafting with use of at least one ITA were enrolled and matched to 49 comparable nonirradiated individuals by propensity scoring system. Preoperative and postoperative data were collected and compared. A 18-month clinical follow-up was performed. RESULTS: Intraoperative mammary artery flow was assessed by transthoracic Doppler echocardiography probe, and there was no significant difference between irradiated and nonirradiated individuals (36 +/- 8.3 mL/min vs 39 +/- 7.2 mL/min, p = 0.15). The two study groups were also comparable in terms of survival (overall mortality, 2%) and recurrence of angina and perfusion defect at control stress perfusion nuclear scan (p = 0.99 and p = 0.77, respectively). One arterial graft showed stenosis at postoperative angiography. The dose of radiation therapy administered did not correlate with graft flow values after anastomosis. CONCLUSION: Our data suggest that the use of a monolateral or bilateral ITA is not associated with early graft failure in patients with history of chest/mediastinal irradiation. Skeletonization harvesting technique might be recommended because of the frequent presence of fibrous tissue around the in situ vessel. Arterial graft stenosis in thorax-irradiated patients should be attributed to the primary atherosclerotic disease rather than to irradiation itself.

Postoperative atrial fibrillation and total dietary antioxidant capacity in patients undergoing cardiac surgery: The Polyphemus Observational Study.

OBJECTIVE: Postoperative atrial fibrillation is a major cause of morbidity and mortality for stroke after cardiac surgery. Both systemic inflammation and oxidative stress play a role in the initiation of postoperative atrial fibrillation after cardiac surgery. The possible association between long-term intake of antioxidant-rich foods and postoperative atrial fibrillation incidence was examined in patients undergoing cardiac surgery. METHODS: A total of 217 consecutive patients (74% were men; median age, 68.4 years) undergoing cardiac surgery, mainly coronary artery bypass grafting and valve replacement or repair, were recruited from January 2010 to September 2012. Total antioxidant capacity was measured in foods by the Trolox equivalent antioxidant capacity assay. The European Prospective Investigation into Cancer and Nutrition Food Frequency Questionnaire was used for dietary total antioxidant capacity assessment. The association among tertiles of dietary total antioxidant capacity and postoperative atrial fibrillation incidence was assessed using multivariable logistic analysis. RESULTS: The overall incidence of total arrhythmias and postoperative atrial fibrillation was 42.4% and 38.2%, respectively. In multivariable analysis, after adjustment for age, gender, use of hypoglycemic drugs, physical activity, education, previous diagnosis of atrial fibrillation, and total energy intake, patients in the highest tertile of dietary total antioxidant capacity had a lower risk of postoperative atrial fibrillation than patients in the 2 lowest tertiles (odds ratio, 0.46; 95% confidence interval, 0.22-0.95; P = .048). A restricted cubic spline transformation confirmed the nonlinear relationship between total antioxidant capacity (in continuous scale) and postoperative atrial fibrillation (P = .023). When considering only coronary artery bypass grafting, valve replacement/repair, and combined surgeries, the protective effect on postoperative atrial fibrillation of a diet rich in antioxidants was confirmed. CONCLUSIONS: Long-term consumption of antioxidant-rich foods is associated with a reduced incidence of postoperative atrial fibrillation in patients undergoing cardiac surgery.