Association and Predictive Value of Preoperative High-Sensitivity C-Reactive Protein for Postoperative Atrial Fibrillation After Video-Assisted Thoracoscopic Lobectomy: A Cohort Study Using the INSPIRE Database
preprint
OA: closed
CC-BY-4.0
Abstract
Abstract Background Postoperative atrial fibrillation (POAF) is a common and serious complication following video-assisted thoracoscopic surgery (VATS), which affects patients' medical outcomes and quality of life. The predictive value of high-sensitivity C-reactive protein (hs-CRP) as an inflammatory marker for POAF in non-cardiac surgeries remains unclear. This study aims to investigate the association and predictive capability of preoperative hs-CRP levels with postoperative new-onset atrial fibrillation. Methods This study employed a retrospective cohort design, analyzing data from the INSPIRE database, which included 3,219 patients who underwent thoracoscopic lobectomy. Patients were classified into low-risk, moderate-risk, and high-risk groups based on preoperative hs-CRP levels. Multivariable logistic regression models were used to assess the independent association of hs-CRP with the occurrence of POAF in different risk groups. Additionally, the incremental predictive utility of hs-CRP for POAF was evaluated using C-statistics, continuous net reclassification improvement (NRI), and integrated discrimination improvement (IDI), in order to quantify the improvement in predictive accuracy when including hs-CRP in existing risk models. Feature importance calculations and predictive models were generated using the Boruta algorithm and machine learning methods. Results The results indicated that the incidence of POAF increased progressively with rising preoperative hs-CRP levels (2.9% in the low-risk group, 6.0% in the moderate-risk group, and 9.3% in the high-risk group, P < 0.001). After multivariable adjustment, the risk of POAF in the high-risk group remained significantly higher than in the low-risk group (OR 2.03, 95% CI 1.14–3.64, P = 0.017). The inclusion of hs-CRP in the traditional risk model significantly improved the C-statistic from 0.753 to 0.796 (P = 0.003) and exhibited significant NRI (0.371, P < 0.001) and IDI (0.015, P = 0.017). Analysis using machine learning models revealed that hs-CRP was one of the important predictive features among many variables, with the CatBoost model achieving the highest AUC of 0.90. Conclusion Preoperative high-sensitivity C-reactive protein is associated with an increased risk of POAF in patients undergoing thoracoscopic surgery, and its inclusion in clinical risk models can significantly enhance predictive capability.
Full text
163,503 characters
· extracted from
preprint-html
· click to expand
Association and Predictive Value of Preoperative High-Sensitivity C-Reactive Protein for Postoperative Atrial Fibrillation After Video-Assisted Thoracoscopic Lobectomy: A Cohort Study Using the INSPIRE Database | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Association and Predictive Value of Preoperative High-Sensitivity C-Reactive Protein for Postoperative Atrial Fibrillation After Video-Assisted Thoracoscopic Lobectomy: A Cohort Study Using the INSPIRE Database Han Wu¹, Zhipeng Zhou², Long Huang³, Xiyuan Xie¹, Liangcheng Qiu¹, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8847142/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 12 You are reading this latest preprint version Abstract Background Postoperative atrial fibrillation (POAF) is a common and serious complication following video-assisted thoracoscopic surgery (VATS), which affects patients' medical outcomes and quality of life. The predictive value of high-sensitivity C-reactive protein (hs-CRP) as an inflammatory marker for POAF in non-cardiac surgeries remains unclear. This study aims to investigate the association and predictive capability of preoperative hs-CRP levels with postoperative new-onset atrial fibrillation. Methods This study employed a retrospective cohort design, analyzing data from the INSPIRE database, which included 3,219 patients who underwent thoracoscopic lobectomy. Patients were classified into low-risk, moderate-risk, and high-risk groups based on preoperative hs-CRP levels. Multivariable logistic regression models were used to assess the independent association of hs-CRP with the occurrence of POAF in different risk groups. Additionally, the incremental predictive utility of hs-CRP for POAF was evaluated using C-statistics, continuous net reclassification improvement (NRI), and integrated discrimination improvement (IDI), in order to quantify the improvement in predictive accuracy when including hs-CRP in existing risk models. Feature importance calculations and predictive models were generated using the Boruta algorithm and machine learning methods. Results The results indicated that the incidence of POAF increased progressively with rising preoperative hs-CRP levels (2.9% in the low-risk group, 6.0% in the moderate-risk group, and 9.3% in the high-risk group, P < 0.001). After multivariable adjustment, the risk of POAF in the high-risk group remained significantly higher than in the low-risk group (OR 2.03, 95% CI 1.14–3.64, P = 0.017). The inclusion of hs-CRP in the traditional risk model significantly improved the C-statistic from 0.753 to 0.796 (P = 0.003) and exhibited significant NRI (0.371, P < 0.001) and IDI (0.015, P = 0.017). Analysis using machine learning models revealed that hs-CRP was one of the important predictive features among many variables, with the CatBoost model achieving the highest AUC of 0.90. Conclusion Preoperative high-sensitivity C-reactive protein is associated with an increased risk of POAF in patients undergoing thoracoscopic surgery, and its inclusion in clinical risk models can significantly enhance predictive capability. High-sensitivity C-reactive protein New-onset atrial fibrillation Video-assisted thoracoscopic surgery Machine learning Risk assessment. Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction Postoperative atrial fibrillation (POAF) is a common arrhythmic complication of thoracic surgery, with an incidence rate of 10% to 30% following lobectomy[ 1 ]. It severely impacts patient prognosis and medical resource consumption. POAF not only leads to prolonged hospital stays and increased healthcare costs but is also closely associated with a heightened risk of stroke, an increased incidence of long-term cardiovascular events, and a rise in overall mortality[ 2 – 10 ]. Current research suggests that the occurrence of postoperative atrial fibrillation is the result of multiple factors, including intraoperative cardiac manipulation, autonomic nervous system dysfunction, electrolyte imbalances, hypoxemia, and systemic inflammatory responses[ 11 , 12 ]. Among these, the role of inflammatory responses in the development of POAF has received increasing attention. The systemic inflammatory response triggered by surgical trauma can promote atrial electrical and structural remodeling through various pathways[ 11 – 16 ]. Therefore, evaluating the inflammatory status preoperatively or during the perioperative period may provide important evidence for risk stratification and prevention of POAF. High-sensitivity C-reactive protein (hs-CRP) is one of the most widely used systemic inflammatory markers in clinical practice, offering advantages such as ease of detection, low cost, and high standardization. In the cardiovascular field, hs-CRP has been shown to predict the risk of various cardiovascular events, including coronary artery disease and heart failure[ 17 – 21 ]. Numerous studies have found that elevated hs-CRP levels in patients undergoing cardiac surgery, either preoperatively or early postoperatively, are associated with an increased incidence of POAF[ 11 , 22 – 26 ]. However, the value of hs-CRP in predicting postoperative atrial fibrillation following non-cardiac thoracic surgery, especially thoracoscopic surgery, has been relatively underexplored. Consequently, the clinical significance of assessing preoperative hs-CRP as an easily accessible laboratory index for POAF risk evaluation in patients undergoing thoracoscopic surgery is substantial. This study utilizes the INSPIRE database, a large perioperative cohort, to investigate the association between preoperative hs-CRP levels and postoperative new-onset atrial fibrillation, aiming to quantitatively assess the predictive capability of hs-CRP for POAF. Methods Study population This retrospective study utilized perioperative data obtained from INSPIRE (version 1.1), a comprehensive perioperative medical database from academic institutions in Korea. INSPIRE covers the period from 2011 to 2020 and contains high-quality medical records for approximately 130,000 patients who underwent surgery under anesthesia. These records include detailed patient demographics, vital signs, comorbidities, laboratory results from pre-admission to six months after discharge, and in-hospital medication data. Access to and extraction of data from the database were managed by one of the authors, Han Wu, in compliance with all required data access protocols. The study population consisted of patients aged ≥ 18 years who underwent elective VATS lobectomy, identified using the International Classification of Diseases, 10th Revision, Procedure Coding System (ICD-10-PCS). The study was reported in accordance with the STROCSS (Strengthening the Reporting of Cohort Studies in Surgery) guidelines[ 27 ]. The exclusion criteria were as follows: a history of atrial fibrillation, atrial flutter, or other sustained supraventricular tachyarrhythmias prior to surgery; previous cardiac surgery or the presence of moderate to severe valvular heart disease; emergency surgery or concomitant major non-thoracic surgery; intraoperative conversion to pneumonectomy; and missing data on hs-CRP or postoperative atrial fibrillation outcomes. After applying these criteria, a total of 3,219 patients were included in the final analysis. Based on preoperative hs-CRP levels, patients were categorized into three risk groups: low-risk (n = 2,703), intermediate-risk (n = 235), and high-risk (n = 281), with the following cut-off values: low risk < 1 mg/L, intermediate risk 1–3 mg/L, and high risk ≥ 3 mg/L ( Figure S1 ). This stratification was predefined according to previous literature and the distribution characteristics of hs-CRP in this cohort[ 17 , 28 , 29 ]. Ethics statement This study used the publicly available INSPIRE database. Institutional Review Board approval was obtained from Seoul National University Hospital (IRB No. H-2210-078-1368) with a waiver of informed consent due to the retrospective nature of the study, and the dataset was released after review and adequate de-identification by the Institutional Data Review Board (DRB No. BD-R-2022-11-02). This study was conducted in accordance with the principles of the Declaration of Helsinki. Data collection and definitions Data extraction was performed using PostgreSQL (version 13.8), which employs Structured Query Language (SQL) for precise information retrieval. The extracted candidate variables were grouped into five categories: (1) demographic details, such as age, sex, and body mass index (BMI); (2) comorbidities, including diabetes, hypertension, carotid artery stenosis, prior cerebral infarction, coronary heart disease (CHD), chronic obstructive pulmonary disease (COPD), and chronic kidney disease (CKD); (3) cardiovascular medication use, including angiotensin-converting enzyme inhibitors (ACEIs), angiotensin receptor blockers (ARBs), β-blockers, calcium channel blockers (CCBs), dipeptidyl peptidase-4 (DPP-4) inhibitors, metformin, sulfonylureas, thiazolidinediones, insulin, antiplatelet agents, anticoagulants, and statins; (4) laboratory tests, including preoperative hemoglobin, hematocrit, serum albumin, white blood cell (WBC) count, platelet count, serum creatinine, activated partial thromboplastin time (APTT), fibrinogen, blood glucose, lymphocyte percentage, and neutrophil percentage; and (5) surgical details, including the resected lobe location, side of resection, number of lobes resected, American Society of Anesthesiologists (ASA) physical status classification, operative time, anesthesia time, intraoperative hypotension, intraoperative bradycardia, total intraoperative fluid volume, estimated blood loss, and urine output. Operative time was defined as the interval from skin incision to completion of skin closure; anesthesia time was defined as the interval from the start of anesthesia induction to completion of extubation. Intraoperative hypotension (IOH) was defined as a mean arterial pressure below 65 mmHg or a decrease of more than 20% from the preoperative value on two consecutive measurements. Intraoperative bradycardia was defined as a heart rate below 50 bpm on two consecutive measurements[ 30 , 31 ]. All laboratory variables were extracted from the first tests performed after hospital admission. Clinical outcomes The primary clinical outcome of this study was postoperative atrial fibrillation (POAF). POAF was defined as a new-onset episode of atrial fibrillation occurring between the end of surgery and hospital discharge, confirmed by electrocardiography, with a duration of ≥ 30 seconds or necessitating therapeutic intervention, including pharmacological heart rate/rhythm control, electrical cardioversion, or the initiation of anticoagulation therapy. Patients with a preoperative history of atrial fibrillation or other persistent supraventricular tachycardias were excluded during the enrollment phase and therefore were not counted as POAF cases. This study did not differentiate between paroxysmal, persistent, or permanent atrial fibrillation; all events meeting the aforementioned criteria were classified as POAF. Diagnostic information was obtained from the diagnostic tables within the INSPIRE database and identified via ICD-10 codes. Statistical analysis Patients were stratified according to preoperative hs-CRP levels. Continuous variables were presented as mean ± standard deviation or median (interquartile range) based on data distribution characteristics; categorical variables were expressed as frequencies and percentages. Depending on the variable type, one-way analysis of Mann-Whitney U test or Kruskal-Wallis test was used to compare baseline characteristics of continuous variables across different hs-CRP risk groups (low, intermediate, high); the chi-square test or Fisher's exact test was employed to compare categorical variables. Variables with more than 20% missing data were excluded from the analysis, whereas those with less than 20% missing data were imputed using Multiple Imputation by Chained Equations (MICE) to ensure unbiased estimates. The primary outcome of this study was postoperative atrial fibrillation (POAF) occurring during the index hospitalization. The trend chi-square test was used to evaluate the trend in crude POAF incidence across different hs-CRP groups. Univariate and multivariate logistic regression models were constructed to calculate odds ratios (ORs) and 95% confidence intervals (CIs) for POAF across hs-CRP groups, using the low-risk group as the reference. Three stepwise adjusted models were developed: Model 1: unadjusted (crude model); Model 2: adjusted for demographic characteristics and major clinical variables; Model 3: fully adjusted model, incorporating all clinically relevant covariates or those statistically significant in univariate analysis. Linear trends across groups (P for trend) were tested by entering the ordinal categorical variable as a continuous term in the models. Multiple complementary metrics were used to assess the incremental predictive value of hs-CRP over traditional clinical risk models. First, the discriminative ability of the models was quantified using the C-statistic and its 95% CI, with the DeLong test employed to compare C-statistics between models. Second, the continuous net reclassification improvement (NRI) and integrated discrimination improvement (IDI) were used to evaluate improvements in risk reclassification after adding hs-CRP. Corresponding 95% CIs and P values were obtained via bootstrap resampling. To further explore the relative importance of perioperative variables and the performance of different prediction algorithms, a machine learning analysis framework was implemented. Feature selection was conducted using the Boruta algorithm based on a random forest classifier to assess feature importance. This algorithm compares the importance of each original variable against randomly generated "shadow" attributes. Based on Boruta decision rules, variables were classified as "Confirmed," "Tentative," or "Rejected," with their normalized importance scores visualized via boxplots. For model construction and evaluation, multiple supervised learning algorithms were trained to predict POAF, including CatBoost, Light Gradient Boosting Machine (LGBM), Random Forest, XGBoost, Extra Trees, Logistic Regression, Naive Bayes, Decision Tree, Linear Kernel Support Vector Machine (SVM), and K-Nearest Neighbors (KNN). The dataset was randomly split into training and testing sets, stratified by outcome. Hyperparameter tuning was performed on the training set using cross-validation. Model performance was evaluated on the test set using Receiver Operating Characteristic (ROC) curves and AUC; 95% CIs for AUC were estimated using the bootstrap method. All statistical tests were two-sided, and a P value < 0.05 was considered statistically significant. All statistical analyses were performed using R software version 4.4.2. Results Study population and baseline characteristics A total of 3,219 patients undergoing video-assisted thoracoscopic surgery (VATS) lobectomy were included in this study. Based on preoperative hs-CRP levels, patients were categorized into three groups: Low Risk (n = 2,703), Medium Risk (n = 235), and High Risk (n = 281). The baseline characteristics of the study population are summarized in Table 1 . Table 1 Baseline characteristics of the study population stratified by hs-CRP levels. Variables Low Risk Medium Risk High Risk P value (n = 2703) (n = 235) (n = 281) Hs-CRP, Median (IQR) 0.1 (0.0, 0.2) 1.6 (1.1, 2.2) 6.8 (4.6, 9.8) < 0.001 Demographics Age, Median (IQR) 65.0 (55.0, 70.0) 65.0 (55.0, 70.0) 65.0 (55.0, 70.0) 0.057 Male, n (%) 1435 (53.1) 167 (71.1) 202 (71.9) < 0.001 BMI, Median (IQR) 23.4 (21.5, 25.7) 23.4 (20.8, 25.7) 22.5 (20.8, 25.7) 0.005 Comorbidities, n (%) Diabetes 36 (1.3) 10 (4.3) 7 (2.5) 0.002 Hypertension 144 (5.3) 14 (6) 17 (6) 0.822 Carotid artery stenosis 22 (0.8) 1 (0.4) 3 (1.1) 0.775 Cerebral infarction 34 (1.3) 1 (0.4) 7 (2.5) 0.131 CHD 121 (4.5) 11 (4.7) 11 (3.9) 0.894 COPD 194 (7.2) 12 (5.1) 18 (6.4) 0.454 CKD 17 (0.6) 2 (0.9) 3 (1.1) 0.483 Medication history, n (%) ACEI 5 (0.2) 0 (0) 2 (0.7) 0.197 ARB 61 (2.3) 8 (3.4) 13 (4.6) 0.039 β-blockers 82 (3) 15 (6.4) 8 (2.8) 0.020 CCB 327 (12.1) 35 (14.9) 30 (10.7) 0.328 DPP-4 inhibitors 51 (1.9) 9 (3.8) 4 (1.4) 0.125 Metformin 16 (0.6) 2 (0.9) 1 (0.4) 0.791 Antiplatelet 41 (1.5) 5 (2.1) 6 (2.1) 0.499 Anticoagulants 47 (1.7) 8 (3.4) 21 (7.5) < 0.001 Statins 198 (7.3) 24 (10.2) 20 (7.1) 0.264 sulfonylureas 47 (1.7) 13 (5.5) 3 (1.1) < 0.001 thiazolidinediones 8 (0.3) 1 (0.4) 0 (0) 0.600 Insulin 67 (2.5) 16 (6.8) 8 (2.8) < 0.001 Surgical conditions Surgical lung lobe, n (%) Right upper 681 (25.2) 59 (25.1) 66 (23.5) 0.396 Right lower 523 (19.3) 38 (16.2) 35 (12.5) 0.011 Right middle 192 (7.1) 9 (3.8) 13 (4.6) 0.056 Left upper 495 (18.3) 35 (14.9) 42 (14.9) 0.182 Left lower 324 (12) 20 (8.5) 23 (8.2) 0.057 Complete lobectomy, n (%) Right 369 (13.7) 61 (26) 89 (31.7) 2 19 (0.7) 2 (0.9) 6 (2.1) 0.035 ASA, n (%) < 0.001 <3 2553 (94.5) 206 (87.7) 257 (91.5) ≥3 150 (5.5) 29 (12.3) 24 (8.5) Operating time, Median (IQR) 125.0 (95.0, 165.0) 160.0 (105.0, 225.0) 165.0 (115.0, 235.0) < 0.001 Anesthesia time, Median (IQR) 175.0 (140.0, 220.0) 215.0 (150.0, 280.0) 215.0 (160.0, 290.0) < 0.001 Intraoperative hypotension, n (%) 23 (0.9) 4 (1.7) 3 (1.1) 0.351 Intraoperative bradycardia, n (%) 21 (0.8) 2 (0.9) 4 (1.4) 0.430 Intraoperative fluid, Median (IQR) 800.0 (650.0, 1000.0) 900.0 (650.0, 1150.0) 1000.0 (650.0, 1250.0) < 0.001 Blood loss, Median (IQR) 145.8 (83.3, 239.8) 229.6 (134.2, 393.6) 281.2 (146.7, 437.5) < 0.001 Urine output, Median (IQR) 153.2 (87.1, 244.9) 179.1 (105.3, 306.1) 219.7 (135.0, 411.7) < 0.001 Laboratory tests, median (IQR) HB 13.2 (12.5, 14.2) 12.9 (11.9, 14.2) 12.2 (11.0, 13.7) < 0.001 Hct 39.1 (37.0, 42.0) 39.1 (36.1, 42.0) 37.0 (32.7, 40.4) < 0.001 Albumin 4.1 (4.0, 4.3) 3.9 (3.6, 4.1) 3.6 (3.2, 4.0) < 0.001 WBC 5.9 (5.2, 7.0) 7.0 (5.9, 8.9) 7.9 (5.6, 11.1) < 0.001 Platelet 222.0 (189.0, 258.0) 245.0 (189.0, 291.0) 233.0 (178.0, 291.0) < 0.001 Creatinine 0.8 (0.7, 0.9) 0.8 (0.7, 1.0) 0.8 (0.7, 0.9) 0.301 APTT 30.6 (29.1, 33.0) 31.2 (29.4, 33.0) 30.6 (28.5, 32.5) 0.022 Fibrinogen 301.0 (263.0, 327.0) 397.0 (327.0, 452.0) 357.0 (301.0, 489.0) < 0.001 Glucose 110.0 (96.0, 127.9) 115.0 (96.0, 141.5) 110.0 (96.0, 132.0) 0.019 Lymphocyte 31.3 (24.1, 37.8) 24.1 (18.3, 31.3) 18.3 (12.4, 28.7) < 0.001 Neutrophil 58.6 (53.1, 65.4) 65.4 (58.6, 71.7) 69.6 (58.6, 77.8) < 0.001 Outcomes Hospital day, Median (IQR) 7.0 (5.0, 9.0) 9.0 (7.0, 13.0) 10.0 (8.0, 15.0) < 0.001 POAF, n (%) 78 (2.9) 14 (6) 26 (9.3) < 0.001 There were significant differences in gender distribution among the groups (P < 0.001), with a higher proportion of male patients in the Medium and High Risk groups compared to the Low Risk group. Patients in the High Risk group had a significantly lower body mass index (BMI) ( P = 0.005 ). Although the prevalence of diabetes mellitus was higher in the Medium Risk group, insulin usage was significantly more frequent in the Medium and High Risk groups ( P < 0.001 ). Regarding intraoperative variables, the High Risk group was associated with longer operative and anesthesia times, greater estimated blood loss, and higher urine output (all P < 0.001 ). Preoperative laboratory findings revealed that the High Risk group had significantly lower albumin levels and hemoglobin but higher white blood cell (WBC) counts, neutrophil counts, and fibrinogen levels (all P < 0.001 ), reflecting a heightened inflammatory state. Notably, the incidence of POAF significantly increased with rising risk levels (2.9% vs. 6.0% vs. 9.3%, P < 0.001 ). The High Risk group also experienced a significantly longer postoperative hospital stay ( P < 0.001 ). Association between preoperative hs-CRP risk categories and POAF The crude incidence of POAF according to hs-CRP risk strata is illustrated in Fig. 1A . The frequency of POAF increased stepwise from the Low through the Medium to the High Risk group (2.9% vs. 6.0% vs. 9.3%; ( P for trend < 0.001)). Multivariable logistic regression analyses were then performed to evaluate the independent association between hs-CRP risk categories and POAF (Fig. 1B). In the unadjusted model (Model 1), both the Medium Risk group (OR 2.13, 95% CI 1.19–3.83; ( P = 0.011 )) and the High Risk group (OR 3.43, 95% CI 2.16–5.45; ( P < 0.001 )) were associated with a higher risk of POAF compared with the Low Risk group. After adjustment for potential confounders in Model 2, the associations were slightly attenuated but remained significant for both the Medium Risk group (OR 1.83, 95% CI 1.01–3.30; ( P = 0.046 )) and the High Risk group (OR 3.01, 95% CI 1.87–4.80; ( P < 0.001 )). In the fully adjusted model (Model 3), which included all clinical and perioperative covariates, the High Risk group remained an independent predictor of POAF with an approximately twofold higher risk than the Low Risk group (OR 2.03, 95% CI 1.14–3.64; ( P = 0.017 )). In contrast, the association for the Medium Risk group was further attenuated and no longer statistically significant (OR 1.27, 95% CI 0.65–2.48; ( P = 0.479 )). Subgroup analyses Subgroup analyses were conducted to examine the consistency of the association between hs-CRP and POAF across clinically relevant strata (Fig. 2). Overall, higher hs-CRP was consistently associated with an increased risk of POAF in most subgroups. The positive association was observed in both younger and older patients: the ORs for POAF per one-unit increase in hs-CRP were 1.16 (95% CI 1.09–1.23; ( P < 0.001 )) among patients aged < 65 years and 1.09 (95% CI 1.04–1.15; ( P = 0.001 )) among those aged ≥ 65 years (( P for interaction = 0.182 )). Similar patterns were seen in women (OR 1.17, 95% CI 1.09–1.25; ( P < 0.001 )) and men (OR 1.08, 95% CI 1.03–1.13; ( P = 0.003 ); ( P for interaction = 0.055) ), in patients with BMI < 30 kg/m 2 (OR 1.11, 95% CI 1.06–1.15; ( P < 0.001 )) and those with BMI ≥ 30 kg/m 2 (OR 1.22, 95% CI 1.01–1.48; ( P = 0.040 ); ( P for interaction = 0.323 )), and across different ASA classes, Revised Cardiac Risk Index (RCRI) scores, surgical sites, and the number of lobes resected (all ( P for interaction > 0.05 )). Although the association between hs-CRP and POAF was weakened and became statistically non-significant in patients with ASA ≥ 3, RCRI ≥ 3, or those undergoing resection of ≥ 2 lobes, no significant interaction was detected in any subgroup. These findings indicate that the relationship between preoperative hs-CRP and POAF risk is generally robust across a wide range of clinical subpopulations. Incremental predictive value of hs-CRP beyond clinical risk models To determine whether preoperative hs-CRP adds prognostic information beyond established clinical predictors, we evaluated the improvement in model performance after incorporating hs-CRP into two different risk models (Table 2 ). Table 2 Incremental predictive value of hs-CRP for postoperative atrial fibrillation beyond standard clinical risk factors. C-statistic (95% CI) p value NRI (95% CI) p value IDI (95% CI) p value RCRI 0.753(0.706,0.799) Ref Ref RCRI + Hs-CRP 0.796(0.749,0.843) 0.003 0.371(0.198,0.544) < 0.001 0.015(0.003,0.027) 0.017 Baseline risk model 0.775(0.739,0.812) Ref Ref Baseline risk model + Hs-CRP 0.872(0.840,0.903) < 0.001 0.901(0.722,1.080) < 0.001 0.174(0.128,0.221) < 0.001 The Basic Model includes age, sex, body mass index (BMI), Revised Cardiac Risk Index (RCRI) score, and extent of resection. The Basic Model + hs-CRP includes the variables in the Basic Model plus preoperative hs-CRP levels. Abbreviations: hs-CRP, high-sensitivity C-reactive protein; CI, confidence interval; NRI, net reclassification improvement; IDI, integrated discrimination improvement. When hs-CRP was added to the RCRI score, the C-statistic increased from 0.753 (95% CI 0.706–0.799) to 0.796 (95% CI 0.749–0.843; ( P = 0.003 )) (Fig. 3A). Reclassification analyses showed that the inclusion of hs-CRP significantly improved risk stratification, with a continuous net reclassification improvement (NRI) of 0.371 (95% CI 0.198–0.544; ( P < 0.001 )) and an integrated discrimination improvement (IDI) of 0.015 (95% CI 0.003–0.027; ( P = 0.017 )). Similarly, adding hs-CRP to the Baseline Risk Model (including age, sex, BMI, RCRI, and extent of resection) led to a marked gain in discrimination, with the C-statistic increasing from 0.775 (95% CI 0.739–0.812) to 0.872 (95% CI 0.840–0.903; ( P < 0.001 )) (Fig. 3B). The improvement in reclassification was even more pronounced, with an NRI of 0.901 (95% CI 0.722–1.080; ( P < 0.001 )) and an IDI of 0.174 (95% CI 0.128–0.221; ( P < 0.001 )). Taken together, these data indicate that preoperative hs-CRP substantially enhances the predictive performance of conventional clinical risk models for POAF. Machine learning–based feature selection and model performance To further evaluate the relative contribution of perioperative variables and the predictive performance of different algorithms for POAF, we applied a machine learning approach (Fig. 4). Using the Boruta algorithm based on a random forest classifier, we ranked the importance of all candidate predictors (Fig. 4A). Green boxplots represent features confirmed as important, yellow boxplots indicate tentative features, red boxplots denote rejected features, and blue boxplots correspond to shadow attributes. Several clinical variables, including preoperative inflammatory markers such as hs-CRP, were identified as important predictors of POAF, supporting the central role of systemic inflammation in the development of postoperative atrial fibrillation. We then compared the discrimination performance of multiple machine learning models using receiver operating characteristic (ROC) curves (Fig. 4B). Among all models, the CatBoost classifier achieved the highest performance with an AUC of 0.90, followed by Light Gradient Boosting Machine (AUC = 0.88), Random Forest (AUC = 0.87), Extreme Gradient Boosting (AUC = 0.87), and Extra Trees (AUC = 0.86). Logistic regression also showed good discrimination with an AUC of 0.85, whereas Naive Bayes (AUC = 0.80), Decision Tree (AUC = 0.67), Support Vector Machine with a linear kernel (AUC = 0.57), and K Neighbors classifier (AUC = 0.52) performed less well. These findings suggest that tree-based ensemble methods provide the best predictive accuracy for POAF, while consistently highlighting preoperative hs-CRP as one of the key informative features. Sensitivity analysis Full data analysis indicated that the high-risk hs-CRP group was associated with the risk of POFA occurrence ( Table S3 ). A similar trend was observed in the propensity score-matched cohort. The high-risk hs-CRP group remained significantly associated with an increased risk of POFA occurrence, further confirming the aforementioned findings ( Table S4) . Discussion This study, based on the INSPIRE database of thoracoscopic surgery cohorts, demonstrates an independent association between preoperative high-sensitivity C-reactive protein (hs-CRP) levels and postoperative new-onset atrial fibrillation (POAF). This association remains robust even after adjusting for demographic characteristics, comorbidities, cardiovascular medication use, and perioperative variables. Furthermore, incorporating hs-CRP into the base clinical model enhances the model's discriminative ability and reclassification capacity, while also showing good calibration. This suggests that the preoperative inflammatory burden has quantifiable predictive value for POAF, providing a simple and accessible laboratory basis for preoperative risk stratification and individualized perioperative management. The pathophysiological mechanisms underlying hs-CRP as an inflammatory marker predicting POAF are multifaceted. Firstly, elevated preoperative hs-CRP levels indicate a state of chronic, low-grade systemic inflammation[ 32 ]. This condition may already have potential impacts on the atrial myocardium, leading to structural and electrophysiological remodeling[ 6 , 33 , 34 ]. Inflammation can promote atrial fibrosis and disrupt electrical conduction coupling between cardiomyocytes, providing a structural basis for the formation of reentry circuits[ 34 ]. Secondly, patients with high baseline inflammation may trigger a more severe acute inflammatory cascade response when faced with surgical trauma, referred to as the "second hit." This response results in the massive release of inflammatory cytokines (such as IL-6 and TNF-α), which can directly or indirectly affect the ion channel function of atrial myocytes (particularly potassium and calcium channels), leading to shorter action potential durations and increased dispersion of effective refractory periods, thereby lowering the threshold for atrial fibrillation[ 35 , 36 ]. In coronary artery bypass grafting (CABG) and valve surgeries, the association between CRP and postoperative atrial fibrillation has been widely and deeply studied, with relatively solid evidence supporting this link. Numerous systematic reviews and meta-analyses have confirmed that both preoperative and early postoperative CRP levels are significantly associated with the incidence of atrial fibrillation[ 33 , 37 ]. Studies have shown that elevated preoperative CRP is an independent predictive factor for POAF following cardiac surgery, suggesting that an existing chronic inflammatory state renders the atrial substrate more vulnerable and reduces its tolerance to surgical stress[ 38 , 39 ]. Moreover, peak postoperative CRP levels more effectively reflect an individual's inflammatory response intensity to surgical trauma, as patients experiencing POAF have significantly higher postoperative CRP levels compared to those without atrial fibrillation[ 33 ]. In contrast to cardiac surgery, high-quality evidence regarding the relationship between CRP and POAF in non-cardiac thoracic surgeries is notably lacking. Search results indicate that studies specifically addressing the relationship between preoperative CRP and POAF in lung cancer surgeries have not been found. Although non-cardiac thoracic surgeries also trigger systemic inflammatory responses, the baseline characteristics of such patients and the nature of the surgeries differ from those in cardiac procedures, which complicates the extrapolation of results. Additionally, factors such as hypoxemia and pulmonary vascular pressure changes that occur after lung surgeries may also play significant roles in the induction of atrial fibrillation. Therefore, there is currently a lack of high-quality clinical evidence to support the use of CRP as a routine predictive tool for POAF after non-cardiac thoracic surgeries. This study also has certain limitations. Firstly, being a retrospective analysis, the limited sample size may affect the generalizability of the results. Secondly, the study population comes from a single center, which may restrict the applicability of the findings to a broader patient population. Future research should adopt a multi-center prospective design to validate our findings and enhance the external validity of the results. Moreover, considering additional inflammatory markers and potential biomarkers could provide further insights into the risk assessment of POAF. Our findings have significant implications for clinical practice. By proactively detecting and monitoring preoperative hs-CRP levels, clinicians can better identify high-risk patients, enabling the formulation of personalized management and intervention strategies. This approach not only helps reduce the incidence of postoperative complications but also improves overall patient prognosis. Therefore, incorporating inflammatory markers into preoperative risk assessments may offer valuable references for the prevention and management of POAF. Conclusion This study investigated the predictive value of preoperative high-sensitivity C-reactive protein (hs-CRP) levels for the occurrence of new-onset atrial fibrillation (POAF) in thoracoscopic surgery patients. The results demonstrated that preoperative hs-CRP levels are significant for risk stratification, with a notably increased incidence of POAF in the high-risk group. Machine learning analyses identified hs-CRP as a key predictive feature, significantly enhancing the predictive capability of traditional clinical risk models. In conclusion, preoperative hs-CRP levels can serve as an important indicator for evaluating POAF risk in thoracoscopic surgery patients, providing a scientific basis for individualized prevention strategies in clinical practice. Abbreviations ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; ASA, American Society of Anesthesiologists; CABG, coronary artery bypass grafting; CCB, calcium channel blocker; CHD, coronary heart disease; CKD, chronic kidney disease; CRP, C-reactive protein; IDI, integrated discrimination improvement; IOH, intraoperative hypotension; KNN, k-nearest neighbors; LGBM, light gradient boosting machine; MICE, multiple imputation by chained equations; POAF, postoperative atrial fibrillation; RCRI, Revised Cardiac Risk Index; SVM, support vector machine; VATS, video-assisted thoracoscopic surgery; Declarations Acknowledgements The authors thank all colleagues who provided assistance and support during this study. Data source and ethics statement The data used in this study were extracted from the publicly available INSPIRE database (https://physionet.org/content/inspire/1.3/). Access to the database requires registration and successful completion of the Collaborative Institutional Training Initiative (CITI) program for data use certification. Author contributions HW and ZZ conceived and designed the study. HW, XX, and LCQ contributed to data extraction, data curation, and quality control. LCQ and HW performed the statistical analyses and developed the machine-learning models. LH provided clinical interpretation and domain expertise for thoracic surgery and perioperative outcomes. HW drafted the initial manuscript. ZZ, LH, XX, and LCQ critically revised the manuscript for important intellectual content. XW supervised the project, provided methodological oversight, and finalized the manuscript. All authors read and approved the final version of the manuscript. Funding The authors received financial support for the research and/or publication of this article from the National Natural Science Foundation of China (Grant No. 82271238). Data availability No datasets were generated or analysed during the current study. Ethics approval and consent to participate This study was approved by the Institutional Review Board of Seoul National University Hospital (IRB No. H-2210-078-1368), which waived the requirement for informed consent due to the retrospective design. The Institutional Data Review Board approved the public release of the de-identified dataset (DRB No. BD-R-2022-11-02). The study was conducted in accordance with the principles of the Declaration of Helsinki. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Author details ¹ Department of Anesthesiology, Fuzhou University Affiliated Provincial Hospital, Fujian Provincial Hospital, Shengli Clinical Medical College of Fujian Medical University, Fuzhou, China. ² Department of Urology, Fuzhou University Affiliated Provincial Hospital, Fujian Provincial Hospital, Shengli Clinical Medical College of Fujian Medical University, Fuzhou, China. ³ Department of Respiratory and Critical Care Medicine, Fuzhou University Affiliated Provincial Hospital, Fujian Provincial Hospital, Shengli Clinical Medical College of Fujian Medical University, Fuzhou, China. Correspondence: Xiaodan Wu, ¹ Department of Anesthesiology, Fuzhou University Affiliated Provincial Hospital, Fujian Provincial Hospital, Shengli Clinical Medical College of Fujian Medical University, Fuzhou, China. Email: [email protected] . References Crispi V, Isaac E, Abah U, Shackcloth M, Lopez E, Eadington T, et al. Surgical factors associated with new-onset postoperative atrial fibrillation after lung resection: the EPAFT multicentre study. Postgrad Med J. 2022;98:177–82. https://doi.org/10.1136/postgradmedj-2020-138904 . Karacan MN, Yafasova A, Fosbol EL, Tas A, Al-Chaer K, Gundlund A, et al. Long-term risk of heart failure in patients with postoperative atrial fibrillation following non-cardiac surgery: Insights from a nationwide cohort. Eur J Heart Fail. 2025;27:264–74. https://doi.org/10.1002/ejhf.3518 . Bidar E, Zeemering S, Gilbers M, Isaacs A, Verheule S, Zink MD, et al. Clinical and electrophysiological predictors of device-detected new-onset atrial fibrillation during 3 years after cardiac surgery. Europace. 2021;23:1922–30. https://doi.org/10.1093/europace/euab136 . Abdelmoneim SS, Rosenberg E, Meykler M, Patel B, Reddy B, Ho J, et al. The Incidence and Natural Progression of New-Onset Postoperative Atrial Fibrillation. JACC Clin Electrophysiol. 2021;7:1134–44. https://doi.org/10.1016/j.jacep.2021.02.005 . Gilbers MD, Kawczynski MJ, Bidar E, Maesen B, Isaacs A, Winters J, et al. Determinants and impact of postoperative atrial fibrillation burden during 2.5 years of continuous rhythm monitoring after cardiac surgery: Results from the RACE V prospective cohort study. Heart Rhythm. 2025;22:647–60. https://doi.org/10.1016/j.hrthm.2024.08.014 . Spicer S, Hasheminia A, Kandi S, Abu-Omar Y, Fernandez AL, El-Diasty M. Cardiac and pericardial inflammatory changes and post-cardiac surgery atrial fibrillation. Cytokine Growth Factor Rev. 2025;85:158–64. https://doi.org/10.1016/j.cytogfr.2025.07.006 . Albini A, Malavasi VL, Vitolo M, Imberti JF, Marietta M, Lip GYH, et al. Long-term outcomes of postoperative atrial fibrillation following non cardiac surgery: A systematic review and metanalysis. Eur J Intern Med. 2021;85:27–33. https://doi.org/10.1016/j.ejim.2020.12.018 . Ruhlmann F, Tichelbacker T, Mackert AF, Engelhardt D, Leha A, Bernhardt M, et al. Incidence, Associated Risk Factors, and Outcomes of Postoperative Arrhythmia After Upper Gastrointestinal Surgery. JAMA Netw Open. 2022;5:e2223225. https://doi.org/10.1001/jamanetworkopen.2022.23225 . Seiffge DJ, Cancelloni V, Raber L, Paciaroni M, Metzner A, Kirchhof P, et al. Secondary stroke prevention in people with atrial fibrillation: treatments and trials. Lancet Neurol. 2024;23:404–17. https://doi.org/10.1016/S1474-4422(24)00037-1 . Luo J, Qin X, Yuan Y, Zhang Y, Liu J, Wang Y, et al. Association of Atrial Fibrillation Burden With Cardiovascular Outcomes in New-Onset Atrial Fibrillation Complicating Myocardial Infarction. J Am Heart Assoc. 2025;14:e039547. https://doi.org/10.1161/JAHA.124.039547 . Fatehi Hassanabad A, Deniset JF, Fedak PWM. Pericardial Inflammatory Mediators That Can Drive Postoperative Atrial Fibrillation in Cardiac Surgery Patients. Can J Cardiol. 2023;39:1090–102. https://doi.org/10.1016/j.cjca.2023.06.003 . Gaudino M, Di Franco A, Rong LQ, Piccini J, Mack M. Postoperative atrial fibrillation: from mechanisms to treatment. Eur Heart J. 2023;44:1020–39. https://doi.org/10.1093/eurheartj/ehad019 . Dobrev D, Heijman J, Hiram R, Li N, Nattel S. Inflammatory signalling in atrial cardiomyocytes: a novel unifying principle in atrial fibrillation pathophysiology. Nat Rev Cardiol. 2023;20:145–67. https://doi.org/10.1038/s41569-022-00759-w . Richter B, Koller L, Hofer F, Kazem N, Hammer A, Silbert BI, et al. Galectin-3 is an independent predictor of postoperative atrial fibrillation and survival after elective cardiac surgery. Heart Rhythm. 2022;19:1774–80. https://doi.org/10.1016/j.hrthm.2022.06.019 . Le Quilliec E, LeBlanc C-A, Neuilly O, Xiao J, Younes R, Altuntas Y, et al. Atrial cardiomyocytes contribute to the inflammatory status associated with atrial fibrillation in right heart disease. Europace. 2024;26. https://doi.org/10.1093/europace/euae082 . Parent S, Vaka R, Risha Y, Ngo C, Kanda P, Nattel S, et al. Prevention of atrial fibrillation after open-chest surgery with extracellular vesicle therapy. JCI Insight. 2023;8. https://doi.org/10.1172/jci.insight.163297 . Ridker PM, Lei L, Louie MJ, Haddad T, Nicholls SJ, Lincoff AM, et al. Inflammation and Cholesterol as Predictors of Cardiovascular Events Among 13 970 Contemporary High-Risk Patients With Statin Intolerance. Circulation. 2024;149:28–35. https://doi.org/10.1161/CIRCULATIONAHA.123.066213 . Mensah GA, Arnold N, Prabhu SD, Ridker PM, Welty FK. Inflammation and Cardiovascular Disease: 2025 ACC Scientific Statement: A Report of the American College of Cardiology. J Am Coll Cardiol. 2025. https://doi.org/10.1016/j.jacc.2025.08.047 . :S0735-1097(25)07555-2. Hamid A, Yimer WK, Oshunbade AA, Khan MS, Kamimura D, Kipchumba RK, et al. Trajectory of C-Reactive Protein and Incident Heart Failure in Black Adults: The Jackson Heart Study. Circ Heart Fail. 2024;17:e011199. https://doi.org/10.1161/CIRCHEARTFAILURE.123.011199 . Wu I-C, Liu C-S, Cheng W-L, Lin T-T, Chen H-L, Chen P-F, et al. Association of leukocyte mitochondrial DNA copy number with longitudinal C-reactive protein levels and survival in older adults: a cohort study. Immun Ageing. 2022;19:62. https://doi.org/10.1186/s12979-022-00322-8 . McCabe JJ, Walsh C, Gorey S, Harris K, Hervella P, Iglesias-Rey R, et al. C-Reactive Protein, Interleukin-6, and Vascular Recurrence According to Stroke Subtype: An Individual Participant Data Meta-Analysis. Neurology. 2024;102:e208016. https://doi.org/10.1212/WNL.0000000000208016 . Liblik K, Zucker J, Baranchuk A, Fernandez AL, Zhang S, Diasty ME. The role of pericardial fluid biomarkers in predicting post-operative atrial fibrillation, a comprehensive review of current literature. Trends Cardiovasc Med. 2024;34:244–7. https://doi.org/10.1016/j.tcm.2023.02.009 . Heo RH, Wang MK, Meyre PB, Birchenough L, Park L, Vuong K, et al. Associations of Inflammatory Biomarkers With the Risk of Morbidity and Mortality After Cardiac Surgery: A Systematic Review and Meta-analysis. Can J Cardiol. 2023;39:1686–94. https://doi.org/10.1016/j.cjca.2023.07.021 . Fischer MA, Howard-Quijano K, Zong NC, Youn JY, Liu NM, Scovotti J, et al. Perioperative Changes in Plasma Nitrite and IL-6 Levels Predict Postoperative Atrial Fibrillation (POAF) and Acute Kidney Injury (AKI) after Cardiac Surgery. Antioxid (Basel). 2024;13:971. https://doi.org/10.3390/antiox13080971 . Amar D, Zhang H, Chung MK, Tan KS, Desiderio D, Park BJ, et al. Amiodarone with or without N-Acetylcysteine for the Prevention of Atrial Fibrillation after Thoracic Surgery: A Double-blind, Randomized Trial. Anesthesiology. 2022;136:916–26. https://doi.org/10.1097/ALN.0000000000004201 . Yang Y, Du Z, Fang M, Ma Y, Liu Y, Wang T, et al. Metabolic signatures in pericardial fluid and serum are associated with new-onset atrial fibrillation after isolated coronary artery bypass grafting. Transl Res. 2023;256:30–40. https://doi.org/10.1016/j.trsl.2023.01.001 . Rashid R, Sohrabi C, Kerwan A, Franchi T, Mathew G, Nicola M, et al. The STROCSS 2024 guideline: strengthening the reporting of cohort, cross-sectional, and case-control studies in surgery. Int J Surg. 2024;110:3151–65. https://doi.org/10.1097/JS9.0000000000001268 . Zeng G, Zhang C, Song Y, Zhang Z, Xu J, Liu Z, et al. The potential impact of inflammation on the lipid paradox in patients with acute myocardial infarction: a multicenter study. BMC Med. 2024;22:599. https://doi.org/10.1186/s12916-024-03823-z . Perin EC, Borow KM, Henry TD, Mendelsohn FO, Miller LW, Swiggum E, et al. Randomized Trial of Targeted Transendocardial Mesenchymal Precursor Cell Therapy in Patients With Heart Failure. J Am Coll Cardiol. 2023;81:849–63. https://doi.org/10.1016/j.jacc.2022.11.061 . Hallqvist L, Granath F, Huldt E, Bell M. Intraoperative hypotension is associated with acute kidney injury in noncardiac surgery: An observational study. Eur J Anaesthesiol. 2018;35:273–9. https://doi.org/10.1097/EJA.0000000000000735 . Saugel B, Reese PC, Sessler DI, Burfeindt C, Nicklas JY, Pinnschmidt HO, et al. Automated Ambulatory Blood Pressure Measurements and Intraoperative Hypotension in Patients Having Noncardiac Surgery with General Anesthesia: A Prospective Observational Study. Anesthesiology. 2019;131:74–83. https://doi.org/10.1097/ALN.0000000000002703 . Wazni O, Martin DO, Marrouche NF, Shaaraoui M, Chung MK, Almahameed S, et al. C reactive protein concentration and recurrence of atrial fibrillation after electrical cardioversion. Heart. 2005;91:1303–5. https://doi.org/10.1136/hrt.2004.038661 . Weymann A, Popov A-F, Sabashnikov A, Ali-Hasan-Al-Saegh S, Ryazanov M, Tse G, et al. Baseline and postoperative levels of C-reactive protein and interleukins as inflammatory predictors of atrial fibrillation following cardiac surgery: a systematic review and meta-analysis. Kardiol Pol. 2018;76:440–51. https://doi.org/10.5603/KP.a2017.0242 . Maesen B, Nijs J, Maessen J, Allessie M, Schotten U. Post-operative atrial fibrillation: a maze of mechanisms. Europace. 2012;14:159–74. https://doi.org/10.1093/europace/eur208 . Liao J, Zhang S, Yang S, Lu Y, Lu K, Wu Y, et al. Interleukin-6-Mediated-Ca(2+) Handling Abnormalities Contributes to Atrial Fibrillation in Sterile Pericarditis Rats. Front Immunol. 2021;12:758157. https://doi.org/10.3389/fimmu.2021.758157 . Aksoy F, Uysal D, Ibrişim E. Relationship between c-reactive protein/albumin ratio and new-onset atrial fibrillation after coronary artery bypass grafting. Rev Assoc Med Bras (1992). 2020;66:1070–6. https://doi.org/10.1590/1806-9282.66.8.1070 Semeraro GC, Meroni CA, Cipolla CM, Cardinale DM. Atrial Fibrillation after Lung Cancer Surgery: Prediction, Prevention and Anticoagulation Management. Cancers (Basel). 2021;13:4012. https://doi.org/10.3390/cancers13164012 . Kinoshita T, Asai T, Takashima N, Hosoba S, Suzuki T, Kambara A, et al. Preoperative C-reactive protein and atrial fibrillation after off-pump coronary bypass surgery. Eur J Cardiothorac Surg. 2011;40:1298–303. https://doi.org/10.1016/j.ejcts.2011.03.027 . Todorov H, Janssen I, Honndorf S, Bause D, Gottschalk A, Baasner S, et al. Clinical significance and risk factors for new onset and recurring atrial fibrillation following cardiac surgery - a retrospective data analysis. BMC Anesthesiol. 2017;17:163. https://doi.org/10.1186/s12871-017-0455-7 . Additional Declarations No competing interests reported. Supplementary Files SupplementaryMaterials.docx Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 01 Apr, 2026 Reviews received at journal 11 Mar, 2026 Reviews received at journal 04 Mar, 2026 Reviewers agreed at journal 03 Mar, 2026 Reviewers agreed at journal 03 Mar, 2026 Reviews received at journal 28 Feb, 2026 Reviewers agreed at journal 20 Feb, 2026 Reviewers invited by journal 18 Feb, 2026 Editor assigned by journal 18 Feb, 2026 Editor invited by journal 16 Feb, 2026 Submission checks completed at journal 13 Feb, 2026 First submitted to journal 13 Feb, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8847142","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":594667096,"identity":"587c77a0-c0cc-4251-aedf-69db96076b2f","order_by":0,"name":"Han Wu¹","email":"","orcid":"","institution":"Fuzhou University Affiliated Provincial Hospital, Fujian Provincial Hospital, Shengli Clinical Medical College of Fujian Medical University","correspondingAuthor":false,"prefix":"","firstName":"Han","middleName":"","lastName":"Wu¹","suffix":""},{"id":594667097,"identity":"6141212f-784c-4d12-970a-375a1780a55f","order_by":1,"name":"Zhipeng Zhou²","email":"","orcid":"","institution":"Fuzhou University Affiliated Provincial Hospital, Fujian Provincial Hospital, Shengli Clinical Medical College of Fujian Medical University","correspondingAuthor":false,"prefix":"","firstName":"Zhipeng","middleName":"","lastName":"Zhou²","suffix":""},{"id":594667098,"identity":"7311e5de-f349-42c1-868b-55bb1f499d88","order_by":2,"name":"Long Huang³","email":"","orcid":"","institution":"Fuzhou University Affiliated Provincial Hospital, Fujian Provincial Hospital, Shengli Clinical Medical College of Fujian Medical University","correspondingAuthor":false,"prefix":"","firstName":"Long","middleName":"","lastName":"Huang³","suffix":""},{"id":594667099,"identity":"73ef9974-1658-499f-b997-afe1115262c9","order_by":3,"name":"Xiyuan Xie¹","email":"","orcid":"","institution":"Fuzhou University Affiliated Provincial Hospital, Fujian Provincial Hospital, Shengli Clinical Medical College of Fujian Medical University","correspondingAuthor":false,"prefix":"","firstName":"Xiyuan","middleName":"","lastName":"Xie¹","suffix":""},{"id":594667100,"identity":"5c1e6029-16df-4b90-96ad-44805943d965","order_by":4,"name":"Liangcheng Qiu¹","email":"","orcid":"","institution":"Fuzhou University Affiliated Provincial Hospital, Fujian Provincial Hospital, Shengli Clinical Medical College of Fujian Medical University","correspondingAuthor":false,"prefix":"","firstName":"Liangcheng","middleName":"","lastName":"Qiu¹","suffix":""},{"id":594667101,"identity":"6d2a1679-6579-4977-bf7e-7e2a7a2d1043","order_by":5,"name":"Xiaodan Wu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAwElEQVRIiWNgGAWjYFACxgYGBjYGBn4wgyQtkg3EawEBoBaDA8QqNriR3PjhQ5ldnvH5w20PfjDYyekSsszgzMFmyRnnkovNbiS2G/YwJBubEbLO7HhjgzRvG3PithuMbRI8DAcStxHUcpix+TdvW33i5v6DbZJ/iNJyvLENaMvhxA0MiW3SRNlif+Zgm+WMc8cTZ9wAapExIMIvkjPSH9/4UFad2N9//Jnkmwo7OYJa0IABacpHwSgYBaNgFOAAAGBKRQpO8dJqAAAAAElFTkSuQmCC","orcid":"","institution":"Fuzhou University Affiliated Provincial Hospital, Fujian Provincial Hospital, Shengli Clinical Medical College of Fujian Medical University","correspondingAuthor":true,"prefix":"","firstName":"Xiaodan","middleName":"","lastName":"Wu","suffix":""}],"badges":[],"createdAt":"2026-02-11 04:53:36","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8847142/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8847142/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":103325536,"identity":"3f0cf7f8-35cf-459d-b6fd-1d421b8c3fa7","added_by":"auto","created_at":"2026-02-24 12:50:32","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1293721,"visible":true,"origin":"","legend":"\u003cp\u003eLegend not included with this version\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-8847142/v1/9ab5a8bf16f96a8c3c04892d.png"},{"id":103505791,"identity":"76e58b96-a54d-4c8e-9b03-61773ea49222","added_by":"auto","created_at":"2026-02-26 13:33:02","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":370195,"visible":true,"origin":"","legend":"\u003cp\u003eLegend not included with this version\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-8847142/v1/d1744a7b22ba1cfd0965bb68.png"},{"id":103506765,"identity":"ce5314c6-ca21-45c1-9f46-efde04b502eb","added_by":"auto","created_at":"2026-02-26 13:39:23","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1196307,"visible":true,"origin":"","legend":"\u003cp\u003eLegend not included with this version\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-8847142/v1/33f2096bba1d1b0653ec70e4.png"},{"id":104397576,"identity":"21d368dc-0c3c-4069-bdd2-3b88af15ebdd","added_by":"auto","created_at":"2026-03-11 11:52:23","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":3020124,"visible":true,"origin":"","legend":"\u003cp\u003eLegend not included with this version\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-8847142/v1/d4e3fe1eca65d2cc6bd95645.png"},{"id":104410112,"identity":"d7a15ca6-0bf6-408e-ad4e-18d7feb8a4da","added_by":"auto","created_at":"2026-03-11 12:49:34","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":7273683,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8847142/v1/5ed62b5e-9b41-44f4-9b32-6f91189a6915.pdf"},{"id":103506567,"identity":"502df233-75d5-4445-a856-f41eee8d5155","added_by":"auto","created_at":"2026-02-26 13:37:42","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":74775,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterials.docx","url":"https://assets-eu.researchsquare.com/files/rs-8847142/v1/57581361b5d8b49dd7c3ad52.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"\u003cp\u003eAssociation and Predictive Value of Preoperative High-Sensitivity C-Reactive Protein for Postoperative Atrial Fibrillation After Video-Assisted Thoracoscopic Lobectomy: A Cohort Study Using the INSPIRE Database\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003ePostoperative atrial fibrillation (POAF) is a common arrhythmic complication of thoracic surgery, with an incidence rate of 10% to 30% following lobectomy[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. It severely impacts patient prognosis and medical resource consumption. POAF not only leads to prolonged hospital stays and increased healthcare costs but is also closely associated with a heightened risk of stroke, an increased incidence of long-term cardiovascular events, and a rise in overall mortality[\u003cspan additionalcitationids=\"CR3 CR4 CR5 CR6 CR7 CR8 CR9\" citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eCurrent research suggests that the occurrence of postoperative atrial fibrillation is the result of multiple factors, including intraoperative cardiac manipulation, autonomic nervous system dysfunction, electrolyte imbalances, hypoxemia, and systemic inflammatory responses[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. Among these, the role of inflammatory responses in the development of POAF has received increasing attention. The systemic inflammatory response triggered by surgical trauma can promote atrial electrical and structural remodeling through various pathways[\u003cspan additionalcitationids=\"CR12 CR13 CR14 CR15\" citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Therefore, evaluating the inflammatory status preoperatively or during the perioperative period may provide important evidence for risk stratification and prevention of POAF.\u003c/p\u003e \u003cp\u003eHigh-sensitivity C-reactive protein (hs-CRP) is one of the most widely used systemic inflammatory markers in clinical practice, offering advantages such as ease of detection, low cost, and high standardization. In the cardiovascular field, hs-CRP has been shown to predict the risk of various cardiovascular events, including coronary artery disease and heart failure[\u003cspan additionalcitationids=\"CR18 CR19 CR20\" citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. Numerous studies have found that elevated hs-CRP levels in patients undergoing cardiac surgery, either preoperatively or early postoperatively, are associated with an increased incidence of POAF[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan additionalcitationids=\"CR23 CR24 CR25\" citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. However, the value of hs-CRP in predicting postoperative atrial fibrillation following non-cardiac thoracic surgery, especially thoracoscopic surgery, has been relatively underexplored.\u003c/p\u003e \u003cp\u003eConsequently, the clinical significance of assessing preoperative hs-CRP as an easily accessible laboratory index for POAF risk evaluation in patients undergoing thoracoscopic surgery is substantial. This study utilizes the INSPIRE database, a large perioperative cohort, to investigate the association between preoperative hs-CRP levels and postoperative new-onset atrial fibrillation, aiming to quantitatively assess the predictive capability of hs-CRP for POAF.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStudy population\u003c/h2\u003e \u003cp\u003e This retrospective study utilized perioperative data obtained from INSPIRE (version 1.1), a comprehensive perioperative medical database from academic institutions in Korea. INSPIRE covers the period from 2011 to 2020 and contains high-quality medical records for approximately 130,000 patients who underwent surgery under anesthesia. These records include detailed patient demographics, vital signs, comorbidities, laboratory results from pre-admission to six months after discharge, and in-hospital medication data. Access to and extraction of data from the database were managed by one of the authors, Han Wu, in compliance with all required data access protocols.\u003c/p\u003e \u003cp\u003eThe study population consisted of patients aged\u0026thinsp;\u0026ge;\u0026thinsp;18 years who underwent elective VATS lobectomy, identified using the International Classification of Diseases, 10th Revision, Procedure Coding System (ICD-10-PCS). The study was reported in accordance with the STROCSS (Strengthening the Reporting of Cohort Studies in Surgery) guidelines[\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe exclusion criteria were as follows: a history of atrial fibrillation, atrial flutter, or other sustained supraventricular tachyarrhythmias prior to surgery; previous cardiac surgery or the presence of moderate to severe valvular heart disease; emergency surgery or concomitant major non-thoracic surgery; intraoperative conversion to pneumonectomy; and missing data on hs-CRP or postoperative atrial fibrillation outcomes. After applying these criteria, a total of 3,219 patients were included in the final analysis. Based on preoperative hs-CRP levels, patients were categorized into three risk groups: low-risk (n\u0026thinsp;=\u0026thinsp;2,703), intermediate-risk (n\u0026thinsp;=\u0026thinsp;235), and high-risk (n\u0026thinsp;=\u0026thinsp;281), with the following cut-off values: low risk\u0026thinsp;\u0026lt;\u0026thinsp;1 mg/L, intermediate risk 1\u0026ndash;3 mg/L, and high risk\u0026thinsp;\u0026ge;\u0026thinsp;3 mg/L (\u003cb\u003eFigure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/b\u003e). This stratification was predefined according to previous literature and the distribution characteristics of hs-CRP in this cohort[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eEthics statement\u003c/h3\u003e\n\u003cp\u003eThis study used the publicly available INSPIRE database. Institutional Review Board approval was obtained from Seoul National University Hospital (IRB No. H-2210-078-1368) with a waiver of informed consent due to the retrospective nature of the study, and the dataset was released after review and adequate de-identification by the Institutional Data Review Board (DRB No. BD-R-2022-11-02). This study was conducted in accordance with the principles of the Declaration of Helsinki.\u003c/p\u003e\n\u003ch3\u003eData collection and definitions\u003c/h3\u003e\n\u003cp\u003eData extraction was performed using PostgreSQL (version 13.8), which employs Structured Query Language (SQL) for precise information retrieval. The extracted candidate variables were grouped into five categories: (1) demographic details, such as age, sex, and body mass index (BMI); (2) comorbidities, including diabetes, hypertension, carotid artery stenosis, prior cerebral infarction, coronary heart disease (CHD), chronic obstructive pulmonary disease (COPD), and chronic kidney disease (CKD); (3) cardiovascular medication use, including angiotensin-converting enzyme inhibitors (ACEIs), angiotensin receptor blockers (ARBs), β-blockers, calcium channel blockers (CCBs), dipeptidyl peptidase-4 (DPP-4) inhibitors, metformin, sulfonylureas, thiazolidinediones, insulin, antiplatelet agents, anticoagulants, and statins; (4) laboratory tests, including preoperative hemoglobin, hematocrit, serum albumin, white blood cell (WBC) count, platelet count, serum creatinine, activated partial thromboplastin time (APTT), fibrinogen, blood glucose, lymphocyte percentage, and neutrophil percentage; and (5) surgical details, including the resected lobe location, side of resection, number of lobes resected, American Society of Anesthesiologists (ASA) physical status classification, operative time, anesthesia time, intraoperative hypotension, intraoperative bradycardia, total intraoperative fluid volume, estimated blood loss, and urine output. Operative time was defined as the interval from skin incision to completion of skin closure; anesthesia time was defined as the interval from the start of anesthesia induction to completion of extubation. Intraoperative hypotension (IOH) was defined as a mean arterial pressure below 65 mmHg or a decrease of more than 20% from the preoperative value on two consecutive measurements. Intraoperative bradycardia was defined as a heart rate below 50 bpm on two consecutive measurements[\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. All laboratory variables were extracted from the first tests performed after hospital admission.\u003c/p\u003e\n\u003ch3\u003eClinical outcomes\u003c/h3\u003e\n\u003cp\u003eThe primary clinical outcome of this study was postoperative atrial fibrillation (POAF). POAF was defined as a new-onset episode of atrial fibrillation occurring between the end of surgery and hospital discharge, confirmed by electrocardiography, with a duration of \u0026ge;\u0026thinsp;30 seconds or necessitating therapeutic intervention, including pharmacological heart rate/rhythm control, electrical cardioversion, or the initiation of anticoagulation therapy. Patients with a preoperative history of atrial fibrillation or other persistent supraventricular tachycardias were excluded during the enrollment phase and therefore were not counted as POAF cases. This study did not differentiate between paroxysmal, persistent, or permanent atrial fibrillation; all events meeting the aforementioned criteria were classified as POAF. Diagnostic information was obtained from the diagnostic tables within the INSPIRE database and identified via ICD-10 codes.\u003c/p\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eStatistical analysis\u003c/h2\u003e \u003cp\u003ePatients were stratified according to preoperative hs-CRP levels. Continuous variables were presented as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation or median (interquartile range) based on data distribution characteristics; categorical variables were expressed as frequencies and percentages. Depending on the variable type, one-way analysis of Mann-Whitney U test or Kruskal-Wallis test was used to compare baseline characteristics of continuous variables across different hs-CRP risk groups (low, intermediate, high); the chi-square test or Fisher's exact test was employed to compare categorical variables. Variables with more than 20% missing data were excluded from the analysis, whereas those with less than 20% missing data were imputed using Multiple Imputation by Chained Equations (MICE) to ensure unbiased estimates.\u003c/p\u003e \u003cp\u003eThe primary outcome of this study was postoperative atrial fibrillation (POAF) occurring during the index hospitalization. The trend chi-square test was used to evaluate the trend in crude POAF incidence across different hs-CRP groups. Univariate and multivariate logistic regression models were constructed to calculate odds ratios (ORs) and 95% confidence intervals (CIs) for POAF across hs-CRP groups, using the low-risk group as the reference. Three stepwise adjusted models were developed: Model 1: unadjusted (crude model); Model 2: adjusted for demographic characteristics and major clinical variables; Model 3: fully adjusted model, incorporating all clinically relevant covariates or those statistically significant in univariate analysis. Linear trends across groups (P for trend) were tested by entering the ordinal categorical variable as a continuous term in the models. Multiple complementary metrics were used to assess the incremental predictive value of hs-CRP over traditional clinical risk models. First, the discriminative ability of the models was quantified using the C-statistic and its 95% CI, with the DeLong test employed to compare C-statistics between models. Second, the continuous net reclassification improvement (NRI) and integrated discrimination improvement (IDI) were used to evaluate improvements in risk reclassification after adding hs-CRP. Corresponding 95% CIs and P values were obtained via bootstrap resampling.\u003c/p\u003e \u003cp\u003eTo further explore the relative importance of perioperative variables and the performance of different prediction algorithms, a machine learning analysis framework was implemented. Feature selection was conducted using the Boruta algorithm based on a random forest classifier to assess feature importance. This algorithm compares the importance of each original variable against randomly generated \"shadow\" attributes. Based on Boruta decision rules, variables were classified as \"Confirmed,\" \"Tentative,\" or \"Rejected,\" with their normalized importance scores visualized via boxplots.\u003c/p\u003e \u003cp\u003eFor model construction and evaluation, multiple supervised learning algorithms were trained to predict POAF, including CatBoost, Light Gradient Boosting Machine (LGBM), Random Forest, XGBoost, Extra Trees, Logistic Regression, Naive Bayes, Decision Tree, Linear Kernel Support Vector Machine (SVM), and K-Nearest Neighbors (KNN). The dataset was randomly split into training and testing sets, stratified by outcome. Hyperparameter tuning was performed on the training set using cross-validation. Model performance was evaluated on the test set using Receiver Operating Characteristic (ROC) curves and AUC; 95% CIs for AUC were estimated using the bootstrap method.\u003c/p\u003e \u003cp\u003eAll statistical tests were two-sided, and a P value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 was considered statistically significant. All statistical analyses were performed using R software version 4.4.2.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003eStudy population and baseline characteristics\u003c/h2\u003e \u003cp\u003eA total of 3,219 patients undergoing video-assisted thoracoscopic surgery (VATS) lobectomy were included in this study. Based on preoperative hs-CRP levels, patients were categorized into three groups: Low Risk (n\u0026thinsp;=\u0026thinsp;2,703), Medium Risk (n\u0026thinsp;=\u0026thinsp;235), and High Risk (n\u0026thinsp;=\u0026thinsp;281). The baseline characteristics of the study population are summarized in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eBaseline characteristics of the study population stratified by hs-CRP levels.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariables\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLow Risk\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMedium Risk\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eHigh Risk\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eP value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;2703)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;235)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;281)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHs-CRP, Median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.1 (0.0, 0.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.6 (1.1, 2.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e6.8 (4.6, 9.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eDemographics\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge, Median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e65.0 (55.0, 70.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e65.0 (55.0, 70.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e65.0 (55.0, 70.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.057\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMale, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1435 (53.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e167 (71.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e202 (71.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBMI, Median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e23.4 (21.5, 25.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e23.4 (20.8, 25.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e22.5 (20.8, 25.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.005\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eComorbidities, n (%)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDiabetes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e36 (1.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e10 (4.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7 (2.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.002\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHypertension\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e144 (5.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e14 (6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e17 (6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.822\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCarotid artery stenosis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e22 (0.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1 (0.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3 (1.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.775\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCerebral infarction\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e34 (1.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1 (0.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7 (2.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.131\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCHD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e121 (4.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e11 (4.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e11 (3.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.894\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCOPD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e194 (7.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e12 (5.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e18 (6.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.454\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCKD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e17 (0.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2 (0.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3 (1.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.483\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eMedication history, n (%)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eACEI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5 (0.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0 (0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2 (0.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.197\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eARB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e61 (2.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e8 (3.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e13 (4.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.039\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eβ-blockers\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e82 (3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e15 (6.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e8 (2.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.020\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCCB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e327 (12.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e35 (14.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e30 (10.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.328\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDPP-4 inhibitors\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e51 (1.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e9 (3.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4 (1.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.125\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMetformin\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e16 (0.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2 (0.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1 (0.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.791\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAntiplatelet\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e41 (1.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e5 (2.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e6 (2.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.499\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAnticoagulants\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e47 (1.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e8 (3.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e21 (7.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eStatins\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e198 (7.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e24 (10.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e20 (7.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.264\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003esulfonylureas\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e47 (1.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e13 (5.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3 (1.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ethiazolidinediones\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e8 (0.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1 (0.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0 (0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.600\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eInsulin\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e67 (2.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e16 (6.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e8 (2.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSurgical conditions\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSurgical lung lobe, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRight upper\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e681 (25.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e59 (25.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e66 (23.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.396\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRight lower\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e523 (19.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e38 (16.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e35 (12.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.011\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRight middle\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e192 (7.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e9 (3.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e13 (4.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.056\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLeft upper\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e495 (18.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e35 (14.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e42 (14.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.182\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLeft lower\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e324 (12)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e20 (8.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e23 (8.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.057\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eComplete lobectomy, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e369 (13.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e61 (26)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e89 (31.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e19 (0.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2 (0.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3 (1.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.648\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNumber of lobes resected, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e=2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e81 (3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e9 (3.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4 (1.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.227\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u0026gt;2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e19 (0.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2 (0.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e6 (2.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.035\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eASA, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u0026lt;3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2553 (94.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e206 (87.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e257 (91.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u0026ge;3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e150 (5.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e29 (12.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e24 (8.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOperating time, Median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e125.0 (95.0, 165.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e160.0 (105.0, 225.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e165.0 (115.0, 235.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAnesthesia time, Median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e175.0 (140.0, 220.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e215.0 (150.0, 280.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e215.0 (160.0, 290.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIntraoperative hypotension, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e23 (0.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4 (1.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3 (1.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.351\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIntraoperative bradycardia, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e21 (0.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2 (0.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4 (1.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.430\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIntraoperative fluid, Median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e800.0 (650.0, 1000.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e900.0 (650.0, 1150.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1000.0 (650.0, 1250.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBlood loss, Median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e145.8 (83.3, 239.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e229.6 (134.2, 393.6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e281.2 (146.7, 437.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUrine output, Median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e153.2 (87.1, 244.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e179.1 (105.3, 306.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e219.7 (135.0, 411.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eLaboratory tests, median (IQR)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHB\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e13.2 (12.5, 14.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e12.9 (11.9, 14.2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e12.2 (11.0, 13.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHct\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e39.1 (37.0, 42.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e39.1 (36.1, 42.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e37.0 (32.7, 40.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAlbumin\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4.1 (4.0, 4.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.9 (3.6, 4.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.6 (3.2, 4.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWBC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5.9 (5.2, 7.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7.0 (5.9, 8.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7.9 (5.6, 11.1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePlatelet\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e222.0 (189.0, 258.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e245.0 (189.0, 291.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e233.0 (178.0, 291.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCreatinine\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.8 (0.7, 0.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.8 (0.7, 1.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.8 (0.7, 0.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.301\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAPTT\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e30.6 (29.1, 33.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e31.2 (29.4, 33.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e30.6 (28.5, 32.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.022\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFibrinogen\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e301.0 (263.0, 327.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e397.0 (327.0, 452.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e357.0 (301.0, 489.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGlucose\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e110.0 (96.0, 127.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e115.0 (96.0, 141.5)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e110.0 (96.0, 132.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.019\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLymphocyte\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e31.3 (24.1, 37.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e24.1 (18.3, 31.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e18.3 (12.4, 28.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNeutrophil\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e58.6 (53.1, 65.4)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e65.4 (58.6, 71.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e69.6 (58.6, 77.8)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eOutcomes\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHospital day, Median (IQR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e7.0 (5.0, 9.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e9.0 (7.0, 13.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e10.0 (8.0, 15.0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePOAF, n (%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e78 (2.9)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e14 (6)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e26 (9.3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThere were significant differences in gender distribution among the groups (P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), with a higher proportion of male patients in the Medium and High Risk groups compared to the Low Risk group. Patients in the High Risk group had a significantly lower body mass index (BMI) (\u003cem\u003eP\u0026thinsp;=\u0026thinsp;0.005\u003c/em\u003e). Although the prevalence of diabetes mellitus was higher in the Medium Risk group, insulin usage was significantly more frequent in the Medium and High Risk groups (\u003cem\u003eP\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/em\u003e).\u003c/p\u003e \u003cp\u003eRegarding intraoperative variables, the High Risk group was associated with longer operative and anesthesia times, greater estimated blood loss, and higher urine output (all \u003cem\u003eP\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/em\u003e). Preoperative laboratory findings revealed that the High Risk group had significantly lower albumin levels and hemoglobin but higher white blood cell (WBC) counts, neutrophil counts, and fibrinogen levels (all \u003cem\u003eP\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/em\u003e), reflecting a heightened inflammatory state.\u003c/p\u003e \u003cp\u003eNotably, the incidence of POAF significantly increased with rising risk levels (2.9% vs. 6.0% vs. 9.3%, \u003cem\u003eP\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/em\u003e). The High Risk group also experienced a significantly longer postoperative hospital stay (\u003cem\u003eP\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/em\u003e).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eAssociation between preoperative hs-CRP risk categories and POAF\u003c/h3\u003e\n\u003cp\u003eThe crude incidence of POAF according to hs-CRP risk strata is illustrated in \u003cb\u003eFig.\u0026nbsp;1A\u003c/b\u003e. The frequency of POAF increased stepwise from the Low through the Medium to the High Risk group (2.9% vs. 6.0% vs. 9.3%; (\u003cem\u003eP for trend\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001)).\u003c/p\u003e \u003cp\u003eMultivariable logistic regression analyses were then performed to evaluate the independent association between hs-CRP risk categories and POAF (Fig.\u0026nbsp;1B). In the unadjusted model (Model 1), both the Medium Risk group (OR 2.13, 95% CI 1.19\u0026ndash;3.83; (\u003cem\u003eP\u0026thinsp;=\u0026thinsp;0.011\u003c/em\u003e)) and the High Risk group (OR 3.43, 95% CI 2.16\u0026ndash;5.45; (\u003cem\u003eP\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/em\u003e)) were associated with a higher risk of POAF compared with the Low Risk group. After adjustment for potential confounders in Model 2, the associations were slightly attenuated but remained significant for both the Medium Risk group (OR 1.83, 95% CI 1.01\u0026ndash;3.30; (\u003cem\u003eP\u0026thinsp;=\u0026thinsp;0.046\u003c/em\u003e)) and the High Risk group (OR 3.01, 95% CI 1.87\u0026ndash;4.80; (\u003cem\u003eP\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/em\u003e)).\u003c/p\u003e \u003cp\u003eIn the fully adjusted model (Model 3), which included all clinical and perioperative covariates, the High Risk group remained an independent predictor of POAF with an approximately twofold higher risk than the Low Risk group (OR 2.03, 95% CI 1.14\u0026ndash;3.64; (\u003cem\u003eP\u0026thinsp;=\u0026thinsp;0.017\u003c/em\u003e)). In contrast, the association for the Medium Risk group was further attenuated and no longer statistically significant (OR 1.27, 95% CI 0.65\u0026ndash;2.48; (\u003cem\u003eP\u0026thinsp;=\u0026thinsp;0.479\u003c/em\u003e)).\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eSubgroup analyses\u003c/h2\u003e \u003cp\u003eSubgroup analyses were conducted to examine the consistency of the association between hs-CRP and POAF across clinically relevant strata (Fig.\u0026nbsp;2). Overall, higher hs-CRP was consistently associated with an increased risk of POAF in most subgroups.\u003c/p\u003e \u003cp\u003eThe positive association was observed in both younger and older patients: the ORs for POAF per one-unit increase in hs-CRP were 1.16 (95% CI 1.09\u0026ndash;1.23; (\u003cem\u003eP\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/em\u003e)) among patients aged\u0026thinsp;\u0026lt;\u0026thinsp;65 years and 1.09 (95% CI 1.04\u0026ndash;1.15; (\u003cem\u003eP\u0026thinsp;=\u0026thinsp;0.001\u003c/em\u003e)) among those aged\u0026thinsp;\u0026ge;\u0026thinsp;65 years ((\u003cem\u003eP for interaction\u0026thinsp;=\u0026thinsp;0.182\u003c/em\u003e)). Similar patterns were seen in women (OR 1.17, 95% CI 1.09\u0026ndash;1.25; (\u003cem\u003eP\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/em\u003e)) and men (OR 1.08, 95% CI 1.03\u0026ndash;1.13; (\u003cem\u003eP\u0026thinsp;=\u0026thinsp;0.003\u003c/em\u003e); (\u003cem\u003eP for interaction\u0026thinsp;=\u0026thinsp;0.055)\u003c/em\u003e), in patients with BMI\u0026thinsp;\u0026lt;\u0026thinsp;30 kg/m\u003csup\u003e2\u003c/sup\u003e (OR 1.11, 95% CI 1.06\u0026ndash;1.15; (\u003cem\u003eP\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/em\u003e)) and those with BMI\u0026thinsp;\u0026ge;\u0026thinsp;30 kg/m\u003csup\u003e2\u003c/sup\u003e (OR 1.22, 95% CI 1.01\u0026ndash;1.48; (\u003cem\u003eP\u0026thinsp;=\u0026thinsp;0.040\u003c/em\u003e); (\u003cem\u003eP for interaction\u0026thinsp;=\u0026thinsp;0.323\u003c/em\u003e)), and across different ASA classes, Revised Cardiac Risk Index (RCRI) scores, surgical sites, and the number of lobes resected (all (\u003cem\u003eP for interaction\u0026thinsp;\u0026gt;\u0026thinsp;0.05\u003c/em\u003e)).\u003c/p\u003e \u003cp\u003eAlthough the association between hs-CRP and POAF was weakened and became statistically non-significant in patients with ASA\u0026thinsp;\u0026ge;\u0026thinsp;3, RCRI\u0026thinsp;\u0026ge;\u0026thinsp;3, or those undergoing resection of \u0026ge;\u0026thinsp;2 lobes, no significant interaction was detected in any subgroup. These findings indicate that the relationship between preoperative hs-CRP and POAF risk is generally robust across a wide range of clinical subpopulations.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eIncremental predictive value of hs-CRP beyond clinical risk models\u003c/h2\u003e \u003cp\u003eTo determine whether preoperative hs-CRP adds prognostic information beyond established clinical predictors, we evaluated the improvement in model performance after incorporating hs-CRP into two different risk models (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eIncremental predictive value of hs-CRP for postoperative atrial fibrillation beyond standard clinical risk factors.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eC-statistic (95% CI)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ep value\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eNRI (95% CI)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ep value\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eIDI (95% CI)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003ep value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRCRI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.753(0.706,0.799)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRCRI\u0026thinsp;+\u0026thinsp;Hs-CRP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.796(0.749,0.843)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.003\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.371(0.198,0.544)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.015(0.003,0.027)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.017\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBaseline risk model\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.775(0.739,0.812)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eRef\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBaseline risk model\u0026thinsp;+\u0026thinsp;Hs-CRP\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.872(0.840,0.903)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.901(0.722,1.080)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.174(0.128,0.221)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"7\"\u003eThe Basic Model includes age, sex, body mass index (BMI), Revised Cardiac Risk Index (RCRI) score, and extent of resection. The Basic Model\u0026thinsp;+\u0026thinsp;hs-CRP includes the variables in the Basic Model plus preoperative hs-CRP levels. Abbreviations: hs-CRP, high-sensitivity C-reactive protein; CI, confidence interval; NRI, net reclassification improvement; IDI, integrated discrimination improvement.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eWhen hs-CRP was added to the RCRI score, the C-statistic increased from 0.753 (95% CI 0.706\u0026ndash;0.799) to 0.796 (95% CI 0.749\u0026ndash;0.843; (\u003cem\u003eP\u0026thinsp;=\u0026thinsp;0.003\u003c/em\u003e)) (Fig.\u0026nbsp;3A). Reclassification analyses showed that the inclusion of hs-CRP significantly improved risk stratification, with a continuous net reclassification improvement (NRI) of 0.371 (95% CI 0.198\u0026ndash;0.544; (\u003cem\u003eP\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/em\u003e)) and an integrated discrimination improvement (IDI) of 0.015 (95% CI 0.003\u0026ndash;0.027; (\u003cem\u003eP\u0026thinsp;=\u0026thinsp;0.017\u003c/em\u003e)).\u003c/p\u003e \u003cp\u003eSimilarly, adding hs-CRP to the Baseline Risk Model (including age, sex, BMI, RCRI, and extent of resection) led to a marked gain in discrimination, with the C-statistic increasing from 0.775 (95% CI 0.739\u0026ndash;0.812) to 0.872 (95% CI 0.840\u0026ndash;0.903; (\u003cem\u003eP\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/em\u003e)) (Fig.\u0026nbsp;3B). The improvement in reclassification was even more pronounced, with an NRI of 0.901 (95% CI 0.722\u0026ndash;1.080; (\u003cem\u003eP\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/em\u003e)) and an IDI of 0.174 (95% CI 0.128\u0026ndash;0.221; (\u003cem\u003eP\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/em\u003e)). Taken together, these data indicate that preoperative hs-CRP substantially enhances the predictive performance of conventional clinical risk models for POAF.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eMachine learning\u0026ndash;based feature selection and model performance\u003c/h2\u003e \u003cp\u003eTo further evaluate the relative contribution of perioperative variables and the predictive performance of different algorithms for POAF, we applied a machine learning approach (Fig.\u0026nbsp;4). Using the Boruta algorithm based on a random forest classifier, we ranked the importance of all candidate predictors (Fig.\u0026nbsp;4A). Green boxplots represent features confirmed as important, yellow boxplots indicate tentative features, red boxplots denote rejected features, and blue boxplots correspond to shadow attributes. Several clinical variables, including preoperative inflammatory markers such as hs-CRP, were identified as important predictors of POAF, supporting the central role of systemic inflammation in the development of postoperative atrial fibrillation.\u003c/p\u003e \u003cp\u003eWe then compared the discrimination performance of multiple machine learning models using receiver operating characteristic (ROC) curves (Fig.\u0026nbsp;4B). Among all models, the CatBoost classifier achieved the highest performance with an AUC of 0.90, followed by Light Gradient Boosting Machine (AUC\u0026thinsp;=\u0026thinsp;0.88), Random Forest (AUC\u0026thinsp;=\u0026thinsp;0.87), Extreme Gradient Boosting (AUC\u0026thinsp;=\u0026thinsp;0.87), and Extra Trees (AUC\u0026thinsp;=\u0026thinsp;0.86). Logistic regression also showed good discrimination with an AUC of 0.85, whereas Naive Bayes (AUC\u0026thinsp;=\u0026thinsp;0.80), Decision Tree (AUC\u0026thinsp;=\u0026thinsp;0.67), Support Vector Machine with a linear kernel (AUC\u0026thinsp;=\u0026thinsp;0.57), and K Neighbors classifier (AUC\u0026thinsp;=\u0026thinsp;0.52) performed less well. These findings suggest that tree-based ensemble methods provide the best predictive accuracy for POAF, while consistently highlighting preoperative hs-CRP as one of the key informative features.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eSensitivity analysis\u003c/h2\u003e \u003cp\u003eFull data analysis indicated that the high-risk hs-CRP group was associated with the risk of POFA occurrence (\u003cb\u003eTable S3\u003c/b\u003e). A similar trend was observed in the propensity score-matched cohort. The high-risk hs-CRP group remained significantly associated with an increased risk of POFA occurrence, further confirming the aforementioned findings (\u003cb\u003eTable S4)\u003c/b\u003e.\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study, based on the INSPIRE database of thoracoscopic surgery cohorts, demonstrates an independent association between preoperative high-sensitivity C-reactive protein (hs-CRP) levels and postoperative new-onset atrial fibrillation (POAF). This association remains robust even after adjusting for demographic characteristics, comorbidities, cardiovascular medication use, and perioperative variables. Furthermore, incorporating hs-CRP into the base clinical model enhances the model's discriminative ability and reclassification capacity, while also showing good calibration. This suggests that the preoperative inflammatory burden has quantifiable predictive value for POAF, providing a simple and accessible laboratory basis for preoperative risk stratification and individualized perioperative management.\u003c/p\u003e \u003cp\u003eThe pathophysiological mechanisms underlying hs-CRP as an inflammatory marker predicting POAF are multifaceted. Firstly, elevated preoperative hs-CRP levels indicate a state of chronic, low-grade systemic inflammation[\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. This condition may already have potential impacts on the atrial myocardium, leading to structural and electrophysiological remodeling[\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. Inflammation can promote atrial fibrosis and disrupt electrical conduction coupling between cardiomyocytes, providing a structural basis for the formation of reentry circuits[\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. Secondly, patients with high baseline inflammation may trigger a more severe acute inflammatory cascade response when faced with surgical trauma, referred to as the \"second hit.\" This response results in the massive release of inflammatory cytokines (such as IL-6 and TNF-α), which can directly or indirectly affect the ion channel function of atrial myocytes (particularly potassium and calcium channels), leading to shorter action potential durations and increased dispersion of effective refractory periods, thereby lowering the threshold for atrial fibrillation[\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eIn coronary artery bypass grafting (CABG) and valve surgeries, the association between CRP and postoperative atrial fibrillation has been widely and deeply studied, with relatively solid evidence supporting this link. Numerous systematic reviews and meta-analyses have confirmed that both preoperative and early postoperative CRP levels are significantly associated with the incidence of atrial fibrillation[\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]. Studies have shown that elevated preoperative CRP is an independent predictive factor for POAF following cardiac surgery, suggesting that an existing chronic inflammatory state renders the atrial substrate more vulnerable and reduces its tolerance to surgical stress[\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]. Moreover, peak postoperative CRP levels more effectively reflect an individual's inflammatory response intensity to surgical trauma, as patients experiencing POAF have significantly higher postoperative CRP levels compared to those without atrial fibrillation[\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eIn contrast to cardiac surgery, high-quality evidence regarding the relationship between CRP and POAF in non-cardiac thoracic surgeries is notably lacking. Search results indicate that studies specifically addressing the relationship between preoperative CRP and POAF in lung cancer surgeries have not been found. Although non-cardiac thoracic surgeries also trigger systemic inflammatory responses, the baseline characteristics of such patients and the nature of the surgeries differ from those in cardiac procedures, which complicates the extrapolation of results. Additionally, factors such as hypoxemia and pulmonary vascular pressure changes that occur after lung surgeries may also play significant roles in the induction of atrial fibrillation. Therefore, there is currently a lack of high-quality clinical evidence to support the use of CRP as a routine predictive tool for POAF after non-cardiac thoracic surgeries.\u003c/p\u003e \u003cp\u003eThis study also has certain limitations. Firstly, being a retrospective analysis, the limited sample size may affect the generalizability of the results. Secondly, the study population comes from a single center, which may restrict the applicability of the findings to a broader patient population. Future research should adopt a multi-center prospective design to validate our findings and enhance the external validity of the results. Moreover, considering additional inflammatory markers and potential biomarkers could provide further insights into the risk assessment of POAF.\u003c/p\u003e \u003cp\u003eOur findings have significant implications for clinical practice. By proactively detecting and monitoring preoperative hs-CRP levels, clinicians can better identify high-risk patients, enabling the formulation of personalized management and intervention strategies. This approach not only helps reduce the incidence of postoperative complications but also improves overall patient prognosis. Therefore, incorporating inflammatory markers into preoperative risk assessments may offer valuable references for the prevention and management of POAF.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis study investigated the predictive value of preoperative high-sensitivity C-reactive protein (hs-CRP) levels for the occurrence of new-onset atrial fibrillation (POAF) in thoracoscopic surgery patients. The results demonstrated that preoperative hs-CRP levels are significant for risk stratification, with a notably increased incidence of POAF in the high-risk group. Machine learning analyses identified hs-CRP as a key predictive feature, significantly enhancing the predictive capability of traditional clinical risk models. In conclusion, preoperative hs-CRP levels can serve as an important indicator for evaluating POAF risk in thoracoscopic surgery patients, providing a scientific basis for individualized prevention strategies in clinical practice.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eACEI, angiotensin-converting enzyme inhibitor;\u003cbr\u003e\u0026nbsp;ARB, angiotensin receptor blocker;\u003cbr\u003e\u0026nbsp;ASA, American Society of Anesthesiologists;\u003cbr\u003e\u0026nbsp;CABG, coronary artery bypass grafting;\u003cbr\u003e\u0026nbsp;CCB, calcium channel blocker;\u003cbr\u003e\u0026nbsp;CHD, coronary heart disease;\u003cbr\u003e\u0026nbsp;CKD, chronic kidney disease;\u003cbr\u003e\u0026nbsp;CRP, C-reactive protein;\u003cbr\u003e\u0026nbsp;IDI, integrated discrimination improvement;\u003cbr\u003e\u0026nbsp;IOH, intraoperative hypotension;\u003cbr\u003e\u0026nbsp;KNN, k-nearest neighbors;\u003cbr\u003e\u0026nbsp;LGBM, light gradient boosting machine;\u003cbr\u003e\u0026nbsp;MICE, multiple imputation by chained equations;\u003cbr\u003e\u0026nbsp;POAF, postoperative atrial fibrillation;\u003cbr\u003e\u0026nbsp;RCRI, Revised Cardiac Risk Index;\u003cbr\u003e\u0026nbsp;SVM, support vector machine;\u003cbr\u003e\u0026nbsp;VATS, video-assisted thoracoscopic surgery;\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors thank all colleagues who provided assistance and support during this study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData source and ethics statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe data used in this study were extracted from the publicly available INSPIRE\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;database (https://physionet.org/content/inspire/1.3/). Access to the database requires registration and successful completion of the Collaborative Institutional Training Initiative (CITI) program for data use certification.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHW and ZZ conceived and designed the study. HW, XX, and LCQ contributed to data extraction, data curation, and quality control. LCQ and HW performed the statistical analyses and developed the machine-learning models. LH provided clinical interpretation and domain expertise for thoracic surgery and perioperative outcomes. HW drafted the initial manuscript. ZZ, LH, XX, and LCQ critically revised the manuscript for important intellectual content. XW supervised the project, provided methodological oversight, and finalized the manuscript. All authors read and approved the final version of the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors received financial support for the research and/or publication of this article from the National Natural Science Foundation of China (Grant No. 82271238).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNo datasets were generated or analysed during the current study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was approved by the Institutional Review Board of Seoul National University Hospital (IRB No. H-2210-078-1368), which waived the requirement for informed consent due to the retrospective design. The Institutional Data Review Board approved the public release of the de-identified dataset (DRB No. BD-R-2022-11-02). The study was conducted in accordance with the principles of the Declaration of Helsinki.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor details\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e¹ Department of Anesthesiology, Fuzhou University Affiliated Provincial Hospital, Fujian Provincial Hospital, Shengli Clinical Medical College of Fujian Medical University, Fuzhou, China.\u003c/p\u003e\n\u003cp\u003e² Department of Urology, Fuzhou University Affiliated Provincial Hospital, Fujian Provincial Hospital, Shengli Clinical Medical College of Fujian Medical University, Fuzhou, China.\u003c/p\u003e\n\u003cp\u003e³ Department of Respiratory and Critical Care Medicine, Fuzhou University Affiliated Provincial Hospital, Fujian Provincial Hospital, Shengli Clinical Medical College of Fujian Medical University, Fuzhou, China.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCorrespondence:\u003c/strong\u003e Xiaodan Wu, ¹ Department of Anesthesiology, Fuzhou University Affiliated Provincial Hospital, Fujian Provincial Hospital, Shengli Clinical Medical College of Fujian Medical University, Fuzhou, China. Email: [email protected].\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eCrispi V, Isaac E, Abah U, Shackcloth M, Lopez E, Eadington T, et al. Surgical factors associated with new-onset postoperative atrial fibrillation after lung resection: the EPAFT multicentre study. Postgrad Med J. 2022;98:177\u0026ndash;82. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1136/postgradmedj-2020-138904\u003c/span\u003e\u003cspan address=\"10.1136/postgradmedj-2020-138904\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKaracan MN, Yafasova A, Fosbol EL, Tas A, Al-Chaer K, Gundlund A, et al. Long-term risk of heart failure in patients with postoperative atrial fibrillation following non-cardiac surgery: Insights from a nationwide cohort. Eur J Heart Fail. 2025;27:264\u0026ndash;74. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/ejhf.3518\u003c/span\u003e\u003cspan address=\"10.1002/ejhf.3518\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBidar E, Zeemering S, Gilbers M, Isaacs A, Verheule S, Zink MD, et al. Clinical and electrophysiological predictors of device-detected new-onset atrial fibrillation during 3 years after cardiac surgery. Europace. 2021;23:1922\u0026ndash;30. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/europace/euab136\u003c/span\u003e\u003cspan address=\"10.1093/europace/euab136\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAbdelmoneim SS, Rosenberg E, Meykler M, Patel B, Reddy B, Ho J, et al. The Incidence and Natural Progression of New-Onset Postoperative Atrial Fibrillation. JACC Clin Electrophysiol. 2021;7:1134\u0026ndash;44. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jacep.2021.02.005\u003c/span\u003e\u003cspan address=\"10.1016/j.jacep.2021.02.005\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGilbers MD, Kawczynski MJ, Bidar E, Maesen B, Isaacs A, Winters J, et al. Determinants and impact of postoperative atrial fibrillation burden during 2.5 years of continuous rhythm monitoring after cardiac surgery: Results from the RACE V prospective cohort study. Heart Rhythm. 2025;22:647\u0026ndash;60. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.hrthm.2024.08.014\u003c/span\u003e\u003cspan address=\"10.1016/j.hrthm.2024.08.014\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSpicer S, Hasheminia A, Kandi S, Abu-Omar Y, Fernandez AL, El-Diasty M. Cardiac and pericardial inflammatory changes and post-cardiac surgery atrial fibrillation. Cytokine Growth Factor Rev. 2025;85:158\u0026ndash;64. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.cytogfr.2025.07.006\u003c/span\u003e\u003cspan address=\"10.1016/j.cytogfr.2025.07.006\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlbini A, Malavasi VL, Vitolo M, Imberti JF, Marietta M, Lip GYH, et al. Long-term outcomes of postoperative atrial fibrillation following non cardiac surgery: A systematic review and metanalysis. Eur J Intern Med. 2021;85:27\u0026ndash;33. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ejim.2020.12.018\u003c/span\u003e\u003cspan address=\"10.1016/j.ejim.2020.12.018\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRuhlmann F, Tichelbacker T, Mackert AF, Engelhardt D, Leha A, Bernhardt M, et al. Incidence, Associated Risk Factors, and Outcomes of Postoperative Arrhythmia After Upper Gastrointestinal Surgery. JAMA Netw Open. 2022;5:e2223225. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1001/jamanetworkopen.2022.23225\u003c/span\u003e\u003cspan address=\"10.1001/jamanetworkopen.2022.23225\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSeiffge DJ, Cancelloni V, Raber L, Paciaroni M, Metzner A, Kirchhof P, et al. Secondary stroke prevention in people with atrial fibrillation: treatments and trials. Lancet Neurol. 2024;23:404\u0026ndash;17. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S1474-4422(24)00037-1\u003c/span\u003e\u003cspan address=\"10.1016/S1474-4422(24)00037-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLuo J, Qin X, Yuan Y, Zhang Y, Liu J, Wang Y, et al. Association of Atrial Fibrillation Burden With Cardiovascular Outcomes in New-Onset Atrial Fibrillation Complicating Myocardial Infarction. J Am Heart Assoc. 2025;14:e039547. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1161/JAHA.124.039547\u003c/span\u003e\u003cspan address=\"10.1161/JAHA.124.039547\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFatehi Hassanabad A, Deniset JF, Fedak PWM. Pericardial Inflammatory Mediators That Can Drive Postoperative Atrial Fibrillation in Cardiac Surgery Patients. Can J Cardiol. 2023;39:1090\u0026ndash;102. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.cjca.2023.06.003\u003c/span\u003e\u003cspan address=\"10.1016/j.cjca.2023.06.003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGaudino M, Di Franco A, Rong LQ, Piccini J, Mack M. Postoperative atrial fibrillation: from mechanisms to treatment. Eur Heart J. 2023;44:1020\u0026ndash;39. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/eurheartj/ehad019\u003c/span\u003e\u003cspan address=\"10.1093/eurheartj/ehad019\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDobrev D, Heijman J, Hiram R, Li N, Nattel S. Inflammatory signalling in atrial cardiomyocytes: a novel unifying principle in atrial fibrillation pathophysiology. Nat Rev Cardiol. 2023;20:145\u0026ndash;67. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41569-022-00759-w\u003c/span\u003e\u003cspan address=\"10.1038/s41569-022-00759-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRichter B, Koller L, Hofer F, Kazem N, Hammer A, Silbert BI, et al. Galectin-3 is an independent predictor of postoperative atrial fibrillation and survival after elective cardiac surgery. Heart Rhythm. 2022;19:1774\u0026ndash;80. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.hrthm.2022.06.019\u003c/span\u003e\u003cspan address=\"10.1016/j.hrthm.2022.06.019\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLe Quilliec E, LeBlanc C-A, Neuilly O, Xiao J, Younes R, Altuntas Y, et al. Atrial cardiomyocytes contribute to the inflammatory status associated with atrial fibrillation in right heart disease. Europace. 2024;26. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/europace/euae082\u003c/span\u003e\u003cspan address=\"10.1093/europace/euae082\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eParent S, Vaka R, Risha Y, Ngo C, Kanda P, Nattel S, et al. Prevention of atrial fibrillation after open-chest surgery with extracellular vesicle therapy. JCI Insight. 2023;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1172/jci.insight.163297\u003c/span\u003e\u003cspan address=\"10.1172/jci.insight.163297\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRidker PM, Lei L, Louie MJ, Haddad T, Nicholls SJ, Lincoff AM, et al. Inflammation and Cholesterol as Predictors of Cardiovascular Events Among 13 970 Contemporary High-Risk Patients With Statin Intolerance. Circulation. 2024;149:28\u0026ndash;35. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1161/CIRCULATIONAHA.123.066213\u003c/span\u003e\u003cspan address=\"10.1161/CIRCULATIONAHA.123.066213\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMensah GA, Arnold N, Prabhu SD, Ridker PM, Welty FK. Inflammation and Cardiovascular Disease: 2025 ACC Scientific Statement: A Report of the American College of Cardiology. J Am Coll Cardiol. 2025. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jacc.2025.08.047\u003c/span\u003e\u003cspan address=\"10.1016/j.jacc.2025.08.047\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. :S0735-1097(25)07555-2.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHamid A, Yimer WK, Oshunbade AA, Khan MS, Kamimura D, Kipchumba RK, et al. Trajectory of C-Reactive Protein and Incident Heart Failure in Black Adults: The Jackson Heart Study. Circ Heart Fail. 2024;17:e011199. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1161/CIRCHEARTFAILURE.123.011199\u003c/span\u003e\u003cspan address=\"10.1161/CIRCHEARTFAILURE.123.011199\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu I-C, Liu C-S, Cheng W-L, Lin T-T, Chen H-L, Chen P-F, et al. Association of leukocyte mitochondrial DNA copy number with longitudinal C-reactive protein levels and survival in older adults: a cohort study. Immun Ageing. 2022;19:62. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12979-022-00322-8\u003c/span\u003e\u003cspan address=\"10.1186/s12979-022-00322-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcCabe JJ, Walsh C, Gorey S, Harris K, Hervella P, Iglesias-Rey R, et al. C-Reactive Protein, Interleukin-6, and Vascular Recurrence According to Stroke Subtype: An Individual Participant Data Meta-Analysis. Neurology. 2024;102:e208016. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1212/WNL.0000000000208016\u003c/span\u003e\u003cspan address=\"10.1212/WNL.0000000000208016\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiblik K, Zucker J, Baranchuk A, Fernandez AL, Zhang S, Diasty ME. The role of pericardial fluid biomarkers in predicting post-operative atrial fibrillation, a comprehensive review of current literature. Trends Cardiovasc Med. 2024;34:244\u0026ndash;7. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.tcm.2023.02.009\u003c/span\u003e\u003cspan address=\"10.1016/j.tcm.2023.02.009\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHeo RH, Wang MK, Meyre PB, Birchenough L, Park L, Vuong K, et al. Associations of Inflammatory Biomarkers With the Risk of Morbidity and Mortality After Cardiac Surgery: A Systematic Review and Meta-analysis. Can J Cardiol. 2023;39:1686\u0026ndash;94. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.cjca.2023.07.021\u003c/span\u003e\u003cspan address=\"10.1016/j.cjca.2023.07.021\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFischer MA, Howard-Quijano K, Zong NC, Youn JY, Liu NM, Scovotti J, et al. Perioperative Changes in Plasma Nitrite and IL-6 Levels Predict Postoperative Atrial Fibrillation (POAF) and Acute Kidney Injury (AKI) after Cardiac Surgery. Antioxid (Basel). 2024;13:971. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/antiox13080971\u003c/span\u003e\u003cspan address=\"10.3390/antiox13080971\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAmar D, Zhang H, Chung MK, Tan KS, Desiderio D, Park BJ, et al. Amiodarone with or without N-Acetylcysteine for the Prevention of Atrial Fibrillation after Thoracic Surgery: A Double-blind, Randomized Trial. Anesthesiology. 2022;136:916\u0026ndash;26. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1097/ALN.0000000000004201\u003c/span\u003e\u003cspan address=\"10.1097/ALN.0000000000004201\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang Y, Du Z, Fang M, Ma Y, Liu Y, Wang T, et al. Metabolic signatures in pericardial fluid and serum are associated with new-onset atrial fibrillation after isolated coronary artery bypass grafting. Transl Res. 2023;256:30\u0026ndash;40. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.trsl.2023.01.001\u003c/span\u003e\u003cspan address=\"10.1016/j.trsl.2023.01.001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRashid R, Sohrabi C, Kerwan A, Franchi T, Mathew G, Nicola M, et al. The STROCSS 2024 guideline: strengthening the reporting of cohort, cross-sectional, and case-control studies in surgery. Int J Surg. 2024;110:3151\u0026ndash;65. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1097/JS9.0000000000001268\u003c/span\u003e\u003cspan address=\"10.1097/JS9.0000000000001268\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZeng G, Zhang C, Song Y, Zhang Z, Xu J, Liu Z, et al. The potential impact of inflammation on the lipid paradox in patients with acute myocardial infarction: a multicenter study. BMC Med. 2024;22:599. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12916-024-03823-z\u003c/span\u003e\u003cspan address=\"10.1186/s12916-024-03823-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePerin EC, Borow KM, Henry TD, Mendelsohn FO, Miller LW, Swiggum E, et al. Randomized Trial of Targeted Transendocardial Mesenchymal Precursor Cell Therapy in Patients With Heart Failure. J Am Coll Cardiol. 2023;81:849\u0026ndash;63. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jacc.2022.11.061\u003c/span\u003e\u003cspan address=\"10.1016/j.jacc.2022.11.061\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHallqvist L, Granath F, Huldt E, Bell M. Intraoperative hypotension is associated with acute kidney injury in noncardiac surgery: An observational study. Eur J Anaesthesiol. 2018;35:273\u0026ndash;9. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1097/EJA.0000000000000735\u003c/span\u003e\u003cspan address=\"10.1097/EJA.0000000000000735\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSaugel B, Reese PC, Sessler DI, Burfeindt C, Nicklas JY, Pinnschmidt HO, et al. Automated Ambulatory Blood Pressure Measurements and Intraoperative Hypotension in Patients Having Noncardiac Surgery with General Anesthesia: A Prospective Observational Study. Anesthesiology. 2019;131:74\u0026ndash;83. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1097/ALN.0000000000002703\u003c/span\u003e\u003cspan address=\"10.1097/ALN.0000000000002703\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWazni O, Martin DO, Marrouche NF, Shaaraoui M, Chung MK, Almahameed S, et al. C reactive protein concentration and recurrence of atrial fibrillation after electrical cardioversion. Heart. 2005;91:1303\u0026ndash;5. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1136/hrt.2004.038661\u003c/span\u003e\u003cspan address=\"10.1136/hrt.2004.038661\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWeymann A, Popov A-F, Sabashnikov A, Ali-Hasan-Al-Saegh S, Ryazanov M, Tse G, et al. Baseline and postoperative levels of C-reactive protein and interleukins as inflammatory predictors of atrial fibrillation following cardiac surgery: a systematic review and meta-analysis. Kardiol Pol. 2018;76:440\u0026ndash;51. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5603/KP.a2017.0242\u003c/span\u003e\u003cspan address=\"10.5603/KP.a2017.0242\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMaesen B, Nijs J, Maessen J, Allessie M, Schotten U. Post-operative atrial fibrillation: a maze of mechanisms. Europace. 2012;14:159\u0026ndash;74. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/europace/eur208\u003c/span\u003e\u003cspan address=\"10.1093/europace/eur208\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiao J, Zhang S, Yang S, Lu Y, Lu K, Wu Y, et al. Interleukin-6-Mediated-Ca(2+) Handling Abnormalities Contributes to Atrial Fibrillation in Sterile Pericarditis Rats. Front Immunol. 2021;12:758157. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fimmu.2021.758157\u003c/span\u003e\u003cspan address=\"10.3389/fimmu.2021.758157\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAksoy F, Uysal D, Ibrişim E. Relationship between c-reactive protein/albumin ratio and new-onset atrial fibrillation after coronary artery bypass grafting. Rev Assoc Med Bras (1992). 2020;66:1070\u0026ndash;6. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1590/1806-9282.66.8.1070\u003c/span\u003e\u003cspan address=\"10.1590/1806-9282.66.8.1070\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSemeraro GC, Meroni CA, Cipolla CM, Cardinale DM. Atrial Fibrillation after Lung Cancer Surgery: Prediction, Prevention and Anticoagulation Management. Cancers (Basel). 2021;13:4012. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/cancers13164012\u003c/span\u003e\u003cspan address=\"10.3390/cancers13164012\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKinoshita T, Asai T, Takashima N, Hosoba S, Suzuki T, Kambara A, et al. Preoperative C-reactive protein and atrial fibrillation after off-pump coronary bypass surgery. Eur J Cardiothorac Surg. 2011;40:1298\u0026ndash;303. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ejcts.2011.03.027\u003c/span\u003e\u003cspan address=\"10.1016/j.ejcts.2011.03.027\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTodorov H, Janssen I, Honndorf S, Bause D, Gottschalk A, Baasner S, et al. Clinical significance and risk factors for new onset and recurring atrial fibrillation following cardiac surgery - a retrospective data analysis. BMC Anesthesiol. 2017;17:163. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12871-017-0455-7\u003c/span\u003e\u003cspan address=\"10.1186/s12871-017-0455-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-anesthesiology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bane","sideBox":"Learn more about [BMC Anesthesiology](http://bmcanesthesiol.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bane","title":"BMC Anesthesiology","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"High-sensitivity C-reactive protein, New-onset atrial fibrillation, Video-assisted thoracoscopic surgery, Machine learning, Risk assessment.","lastPublishedDoi":"10.21203/rs.3.rs-8847142/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8847142/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003ePostoperative atrial fibrillation (POAF) is a common and serious complication following video-assisted thoracoscopic surgery (VATS), which affects patients' medical outcomes and quality of life. The predictive value of high-sensitivity C-reactive protein (hs-CRP) as an inflammatory marker for POAF in non-cardiac surgeries remains unclear. This study aims to investigate the association and predictive capability of preoperative hs-CRP levels with postoperative new-onset atrial fibrillation.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eThis study employed a retrospective cohort design, analyzing data from the INSPIRE database, which included 3,219 patients who underwent thoracoscopic lobectomy. Patients were classified into low-risk, moderate-risk, and high-risk groups based on preoperative hs-CRP levels. Multivariable logistic regression models were used to assess the independent association of hs-CRP with the occurrence of POAF in different risk groups. Additionally, the incremental predictive utility of hs-CRP for POAF was evaluated using C-statistics, continuous net reclassification improvement (NRI), and integrated discrimination improvement (IDI), in order to quantify the improvement in predictive accuracy when including hs-CRP in existing risk models. Feature importance calculations and predictive models were generated using the Boruta algorithm and machine learning methods.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eThe results indicated that the incidence of POAF increased progressively with rising preoperative hs-CRP levels (2.9% in the low-risk group, 6.0% in the moderate-risk group, and 9.3% in the high-risk group, P\u0026thinsp;\u0026lt;\u0026thinsp;0.001). After multivariable adjustment, the risk of POAF in the high-risk group remained significantly higher than in the low-risk group (OR 2.03, 95% CI 1.14\u0026ndash;3.64, P\u0026thinsp;=\u0026thinsp;0.017). The inclusion of hs-CRP in the traditional risk model significantly improved the C-statistic from 0.753 to 0.796 (P\u0026thinsp;=\u0026thinsp;0.003) and exhibited significant NRI (0.371, P\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and IDI (0.015, P\u0026thinsp;=\u0026thinsp;0.017). Analysis using machine learning models revealed that hs-CRP was one of the important predictive features among many variables, with the CatBoost model achieving the highest AUC of 0.90.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003ePreoperative high-sensitivity C-reactive protein is associated with an increased risk of POAF in patients undergoing thoracoscopic surgery, and its inclusion in clinical risk models can significantly enhance predictive capability.\u003c/p\u003e","manuscriptTitle":"Association and Predictive Value of Preoperative High-Sensitivity C-Reactive Protein for Postoperative Atrial Fibrillation After Video-Assisted Thoracoscopic Lobectomy: A Cohort Study Using the INSPIRE Database","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-24 12:50:27","doi":"10.21203/rs.3.rs-8847142/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-04-01T11:54:49+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-11T20:58:40+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-04T06:58:12+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"293961133858662116739896960692631355643","date":"2026-03-04T02:50:48+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"310211425063860462400168743215578274813","date":"2026-03-04T01:38:47+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-28T20:04:55+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"81673471673114404037519276723132272091","date":"2026-02-20T21:48:10+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-02-18T21:36:31+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-02-18T21:34:29+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-02-16T19:36:55+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-02-13T16:57:06+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Anesthesiology","date":"2026-02-13T16:52:38+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-anesthesiology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bane","sideBox":"Learn more about [BMC Anesthesiology](http://bmcanesthesiol.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bane","title":"BMC Anesthesiology","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"dd5fc97b-7825-48a9-a557-0026952f0655","owner":[],"postedDate":"February 24th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-05-06T01:10:03+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-24 12:50:27","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8847142","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8847142","identity":"rs-8847142","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}
Text is read by the "Ask this paper" AI Q&A widget below. Extraction quality varies by source — PMC NXML preserves structure cleanly, OA-HTML may include some navigation residue, and OA-PDF can have broken hyphenation. The publisher copy (via DOI) is the canonical version.
My notes (saved in your browser only)
Ask this paper
Answers must be backed by verbatim quotes from this paper's full text. Hallucinated quotes are dropped automatically; if no verbatim passage answers the question, we say so. How this works
Citation neighborhood (no data yet)
We don't have any in-corpus citations linked to this paper yet. This is a recent paper (2026) — citers typically take a year or two to land, and the OpenAlex reference graph may still be filling in.
Source provenance
- europepmc
- last seen: 2026-05-20T01:45:00.602351+00:00
- unpaywall
- last seen: 2026-05-22T02:00:06.705733+00:00
License: CC-BY-4.0