Identification of biomarkers associated with acute lung injury after cardiopulmonary bypass by integrative transcriptomic analysis and clinical validation

Identification of biomarkers associated with acute lung injury after cardiopulmonary bypass by integrative transcriptomic analysis and clinical validation | 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 Article Identification of biomarkers associated with acute lung injury after cardiopulmonary bypass by integrative transcriptomic analysis and clinical validation Zhicong Liu, Xueke Wang, Xiaohui Zhong, Jiajie Fan, Liyang Ying, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8751380/v1 This work is licensed under a CC BY 4.0 License Status: Under Revision Version 1 posted 14 You are reading this latest preprint version Abstract Pediatric cardiopulmonary bypass (CPB) induces profound systemic inflammation and immune dysregulation that contribute to postoperative complications. However, conserved transcriptional programs underlying CPB-associated immune responses remain incompletely characterized. This study aimed to define reproducible immune transcriptional signatures associated with postoperative acute lung injury in pediatric CPB. Three complementary transcriptomic datasets, including whole-blood bulk RNA sequencing, peripheral blood mononuclear cell single-cell RNA sequencing, and neutrophil bulk RNA sequencing, were integrated to characterize immune responses. Differential expression, functional enrichment, co-expression network analysis, protein-protein interaction analysis, and machine-learning-based feature selection were applied to identify robust genes. Key candidates were validated in pediatric cohorts using quantitative real-time PCR. Cross-dataset integration revealed consistent activation of innate immune and myeloid programs after CPB. Integrated network and machine-learning analyses converged on a three-gene signature comprising CD163 , IL10 , and PPARG . Clinical validation demonstrated significant postoperative upregulation of all three genes, which correlated with the oxygenation index, indicating an association with postoperative ALI severity. This integrative transcriptomic analysis identifies a reproducible three-gene immune signature associated with postoperative acute lung injury following pediatric CPB. These findings provide insight into CPB-induced immune dysregulation and support the potential relevance of compact immune-related gene signatures for early postoperative risk stratification. Health sciences/Biomarkers Biological sciences/Computational biology and bioinformatics Health sciences/Diseases Biological sciences/Genetics Biological sciences/Immunology Health sciences/Medical research Transcriptomics Machine learning Cardiopulmonary bypass Pediatrics Congenital heart disease Acute lung injury Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Introduction Cardiopulmonary bypass (CPB) is widely used in pediatric cardiac surgery including the correction of congenital heart defect [ 1 , 2 ]. However, blood exposure to artificial surfaces, ischemia-reperfusion injury, and hemodilution trigger systemic inflammatory and immune responses that are closely associated with postoperative complications [ 3 – 5 ]. These processes involve innate immune activation, cytokine release, leukocyte recruitment, and endothelial dysfunction [ 6 , 7 ], yet the molecular mechanisms underlying these alterations remain elusive. Although several transcriptomic studies reported CPB-associated changes in whole blood [ 8 – 10 ], leukocyte subsets [ 11 – 13 ], or pediatric myocardial tissue [ 14 ], the available datasets remain sparse, heterogeneous, and typically limited to single cohorts. As a result, the cell-type-resolved and cross-dataset transcriptional architecture of early postoperative immune responses in children remains unknown. Systems-level investigations integrating bulk, single-cell, and compartment-specific transcriptomic layers are particularly scarce. Current clinical biomarkers primarily reflect late inflammatory changes, underscoring the need for early and robust molecular signatures that capture core CPB-induced transcriptional alterations [ 15 ]. To address this gap, we integrated bulk and single-cell transcriptomic datasets from multiple immune compartments to delineate early CPB-induced immune responses and identify reproducible molecular signatures. We further validated these transcriptomic findings in paired pediatric clinical samples and examined their association with postoperative complications and clinical outcomes. Materials and Methods Transcriptomic Datasets Three publicly available transcriptomic datasets profiling human peripheral blood responses to cardiopulmonary bypass (CPB) were analyzed in this study, including whole-blood bulk RNA-seq (GSE143780) [ 8 ], PBMC single-cell RNA-seq (GSE262146) [ 10 ], and neutrophil bulk RNA-seq (GSE297377) [ 13 ]. All raw or processed matrices were obtained directly from the Gene Expression Omnibus (GEO). GSE143780 and GSE297377 were downloaded as count matrices, while GSE262146 was processed from the filtered_feature_bc_matrix.h5 file. For consistency across datasets, the single-cell data were re-processed without reference with the original authors’ annotation. These datasets were selected because they captured comparable perioperative sampling windows and enabled cross-compartment examination of acute transcriptional responses following CPB. Bulk RNA-seq Processing and Differential Expression Analysis Bulk RNA-seq analyses for GSE143780 and GSE297377 were performed in R (version 4.5.2). Gene-level count matrices were normalized and variance-stabilized with the DESeq2 package [ 16 ] (version 1.48.2). Differential expression analysis was conducted using the Wald test implemented in DESeq2, and genes with |log2 fold change| ≥ 1 and adjusted p-value < 0.05 were considered significant. All visualization procedures for the bulk RNA-seq datasets including volcano plots and heatmaps were performed using ggplot2 [ 17 ] (version 4.0.0) and ComplexHeatmap [ 18 ] (version 2.24.0). Single-cell RNA-seq Processing Single-cell data from GSE262146 were processed using Seurat [ 19 ] (version 5.3.1) and SingleCellExperiment [ 20 ] (version 1.30.1). Standard quality-control thresholds were applied to remove cells with low gene counts, excessive mitochondrial percentages, or extreme UMI levels. Normalization, scaling, PCA, clustering, and UMAP embeddings followed the Seurat workflow using the first 20 principal components. Lineage-specific enrichment analyses were conducted using clusterProfiler [ 21 ] (version 4.16.0) and msigdbr [ 22 ] (version 25.1.1). Functional Enrichment Analyses Functional enrichment analyses were conducted with clusterProfiler, ReactomePA [ 23 ] (version 1.52.0), and msigdbr. GO Biological Process and KEGG pathway overrepresentation analyses used DEGs identified from the Wald test. KEGG pathway enrichment analysis was performed using the KEGG database [ 24 ]. Gene Set Enrichment Analysis (GSEA) utilized the full ranked gene list with default parameters (minGSSize = 10, maxGSSize = 500, and 1000 permutations). Because CPB8h exhibited the strongest early transcriptional changes, additional enrichment analyses were performed specifically for this comparison to assess phase-specific pathway activation. Weighted Gene Co-expression Network Analysis Weighted gene co-expression network analysis was conducted on the log2-transformed expression matrix of GSE143780 using the WGCNA package [ 25 ] (version 1.73). A signed hybrid network was constructed after removing low-variance genes. The soft-thresholding power was set to 9 based on scale-free topology criteria. Modules were constructed using blockwiseModules with minModuleSize = 30, mergeCutHeight = 0.25, and default settings otherwise. Module eigengenes were correlated with perioperative CPB phases (PreCPB, CPBend, CPB8h, and CPB24h), and the module showing the strongest association with CPB8h was designated as the CPB-responsive module for downstream analyses. Protein-protein Interaction (PPI) Network Analysis Genes shared across bulk and single-cell DEG intersections were analyzed using STRING [ 26 ] (version 12.0) with default confidence-score thresholds. The interaction network was imported into Cytoscape [ 27 ] (version 3.10.3), and hub genes were ranked using the cytoHubba plugin [ 28 ] via the Maximal Clique Centrality (MCC) algorithm. The top 30 MCC-ranked genes were retained as putative regulatory hubs. All network visualizations were produced directly within Cytoscape using the “circular layout”. Machine-learning-based Gene Selection Machine-learning models were trained using a feature set derived from two sources. First, the top 30 hub genes ranked by the Maximal Clique Centrality (MCC) algorithm in the cytoHubba plugin were extracted from the STRING-derived PPI network. Second, genes belonging to the CPB-responsive WGCNA module from GSE143780 were intersected with genes jointly identified across whole-blood (GSE143780), neutrophil (GSE297377), and PBMC single-cell (GSE262146) differential expression analyses. This intersection set (GeneSet4) represented genes supported by both co-expression structure and multi-compartment transcriptomic reproducibility. The union of the PPI top 30 genes and GeneSet4 formed the feature matrix used for model training. Two regularized logistic regression models—LASSO (α = 1.0) and Elastic Net (0 < α < 1.0)—were implemented with the glmnet package [ 29 ] (version 4.1.10) with optimal λ selected by 10-fold cross-validation. In parallel, a Random Forest classifier was trained using the randomForest package [ 30 ] (version 4.7.1.2) with 1000 trees, and gene importance was quantified using MeanDecreaseGini metrics. Genes consistently prioritized across all three models were considered robust predictors of acute CPB activation and were consolidated into the final three-gene signature. Collecting Patients’ Samples After obtained the approval by the Ethics Committee of Children’s Hospital (Approval No. 2022-IRB-0054), Zhejiang University School of Medicine, Zhejiang, China and written informed consent from their parents, a total of 42 patients under 3 years old who underwent cardiac surgery under CPB were enrolled. Their blood samples were collected into ethylene diamine tetraacetic acid vacuum tubes 24 hours after CPB. The exclusion criteria were as follows: premature infants, those with liver and kidney dysfunction, those with major chromosome abnormalities, those with pulmonary inflammation before surgery, those who required extracorporeal membrane oxygenation support after surgery, those who died because of cardiac dysfunction, or those who refused to provide informed consent. The Second International Guidelines for the Diagnosis and Management of Pediatric Acute Respiratory Distress Syndrome (PALICC-2) [ 31 ] were used to categorize the patients into the ALI group and non-ALI group. In brief, this criterion includes new opacities consistent with acute pulmonary parenchymal disease which are not due primarily to atelectasis or pleural effusion seen on a chest radiograph, an oxygenation index (OI) ≥ 4, and the absence of cardiogenic pulmonary edema. The formula for OI: mean airway pressure (MAP) (cm H 2 O) × FiO 2 /PaO 2 (mm Hg). Quantitative Real-time Polymerase Chain Reaction (qRT-PCR) Genomic DNA was extracted from whole blood samples from the above patient cohort with a commercially available DNA isolation kit (QIAGEN). The obtained cDNA was amplified using TB Green Premix Ex Taq (Takara) on a Roche LightCycler480 PCR System. The expression levels of CD163 , IL10 and PPARG transcripts were normalized to the expression level of ACTB. The gene relative expression was calculated using the 2 −ΔΔCt method. Primer sequences for these genes are listed in Supplementary Table S1 . Statistical Analysis Continuous data were tested for normal distribution with the one-sample Kolmogorov-Smirnov test. Variables were presented as mean values and standard deviations if normally distributed, and otherwise, as median values (interquartile range). The Student’s t test or Mann-Whitney U test were used to determine the significance of variable differences between the two groups where appropriate. A Pearson or Spearman correlation test was performed to determine the correlation between biomarker data and clinical outcomes. For all analyses, a P value less than 0.05 was considered to be statistically significant. Results Timepoint Harmonization across Datasets We analyzed three complementary transcriptomic datasets, including whole-blood bulk RNA-seq (GSE143780) [8], PBMC single-cell RNA-seq (GSE262146) [10], and neutrophil bulk RNA-seq (GSE297377) [13]. An overview of the study design and analytical workflow is presented in Figure 1. To harmonize postoperative timepoints across datasets with different sampling schemes, we first evaluated whether the modified ultrafiltration (MUF) [32,33] procedure in GSE143780 introduced a distinct transcriptional state. A random-gene heatmap and sample-sample correlation analysis showed that CPBend and MUF0h exhibited highly similar expression profiles and did not form separate sample clusters (Supplementary Figure 1a-d), indicating that MUF0h does not represent a transcriptionally distinct state relative to CPBend. Based on this observation, MUF8h and MUF24h were treated as the CPB8h and CPB24h timepoint in subsequent analyses. After aligning all datasets to a unified axis consisting of PreCPB, CPBend (or Post), CPB8h, and CPB24h, we compared early transcriptional responses to CPB across the three transcriptomic datasets. We first summarized the sampling timepoints of whole-blood bulk RNA-seq (GSE143780), PBMC single-cell RNA-seq (GSE262146), and neutrophil bulk RNA-seq (GSE297377) (Figure 2a). Time-course clustering of GSE143780 revealed several distinct temporal expression patterns, including gene groups that were transiently induced, progressively increased, or gradually downregulated after CPB (Figure 2b). Consistently, all three datasets demonstrated widespread differential gene expression when comparing PreCPB and CPBend samples: large sets of upregulated and downregulated genes were detected in GSE143780 (CPB8h vs Pre), GSE262146 (CPB8h vs Pre), and GSE297377 (Post vs Pre), as shown in the heatmaps (Figure 2c-e) and volcano plots (Figure 2f-h). The full differential expression profiles at CPBend and CPB24h in GSE143780, and at Post and CPB24h in GSE262146, are presented for reference (Supplementary Figure 1e-l). Temporal Pathway Dynamics in Whole Blood After CPB To characterize global functional responses to CPB, we performed GOBP and KEGG enrichment analyses across the three postoperative timepoints in GSE143780 (CPBend, CPB8h, and CPB24h). Upregulated pathways were dominated by innate immune and myeloid-related programs, including positive regulation of innate immune response, myeloid leukocyte activation, positive regulation of cytokine production, regulation of inflammatory response and leukocyte chemotaxis. KEGG enrichment further highlighted activation of Toll-like receptor signaling, platelet activation, hematopoietic cell lineage and focal adhesion. In contrast, downregulated GOBP terms were predominantly linked to lymphocyte effector functions, including cell killing, leukocyte-mediated cytotoxicity, leukocyte-mediated immunity and decreased positive regulation of NK cell-mediated immunity and cytotoxicity. KEGG analyses revealed suppression of antigen processing and presentation, along with additional immune-related modules (Figure 3a-d). GSEA was performed for the CPB8h timepoint. Upregulated signatures included monocyte activation, myeloid leukocyte activation, neutrophil degranulation and regulation of chemokine production whereas downregulated signatures involved CD4 + T-cell proliferation, positive regulation of cell killing and NK-cell-mediated cytotoxicity (Figure 3e-f). Lineage-specific Pathway Dynamics in PBMCs after CPB To determine the cellular sources underlying the whole-blood signatures, we examined lineage-specific transcriptional responses using the PBMC single-cell dataset (GSE262146). Major immune lineages including T cells, B cells, NK cells, monocyte-macrophage lineage cells, dendritic cells, progenitors, and cycling cells were annotated using canonical markers (Supplementary Figure 2a). UMAP visualization and compositional changes revealed a marked early expansion of monocyte-macrophage populations after CPB, whereas lymphocyte subsets exhibited a transient postoperative decrease followed by partial recovery at 24 h (Figure 4a-b). Differential expression analyses summarized the sets of upregulated and downregulated genes for each lineage across postoperative timepoints (Figure 4c-d), providing an overview of transcriptional perturbations in individual immune subsets. Monocyte-macrophage lineage cells displayed enrichment of innate and inflammatory programs. Upregulated GOBP terms included myeloid leukocyte activation, regulation of inflammatory response, and response to wounding, while KEGG analysis prominently highlighted complement and coagulation cascades along with additional inflammation-related pathways (Figure 4e; Supplementary Figure 2b). Downregulated pathways in this lineage mapped to broadly shared adaptive immune effector modules, including antigen processing and presentation and T-cell receptor signaling. T cells exhibited modest early upregulation of inflammatory and chemotactic responses, but key adaptive pathways including hematopoietic cell lineage, IL-17 signaling and T-cell activation were downregulated (Supplementary Figure 2c-d). B cells showed consistent reduction of antigen presentation, immune effector processes and broader adaptive immune regulatory modules that are annotated under T-cell receptor signaling and Th1/Th2/Th17 differentiation (Supplementary Figure 2e-k). NK cells demonstrated suppression of cytotoxicity-related pathways and immune-activating receptor signaling (Supplementary Figure 2g). As neutrophils are not captured within the PBMC compartment, we assessed their lineage-specific transcriptional responses using the neutrophil bulk RNA-seq dataset (GSE297377), which contains sorted neutrophil samples and therefore provides a reasonable approximation of lineage-specific signals not captured in PBMCs. Neutrophils exhibited activation of innate immune pathways, including IL-1β production, myeloid leukocyte activation, regulation of innate immune response, and MAPK signaling (Figure 4f-g). Only a small number of pathways, mainly related to apoptosis and leukocyte differentiation, were modestly downregulated. WGCNA Identifies a CPB-responsive Gene Module To identify coordinated gene programs underlying early transcriptional responses to CPB, we performed weighted gene co-expression network analysis (WGCNA) using the whole-blood bulk RNA-seq dataset (GSE143780). Sample clustering revealed no outlier samples across the four timepoints (PreCPB, CPBend, CPB8h, CPB24h), supporting the suitability of the dataset for network construction (Figure 5a). Soft-thresholding analysis identified an appropriate power that achieved scale-free topology while preserving network connectivity (Figure 5b). Hierarchical clustering of module eigengenes and module-module correlation patterns showed distinct co-expression modules with varying degrees of inter-module similarity (Figure 5c-d). Module-trait correlation analysis identified the lightcyan module as the most strongly associated with the CPB8h timepoint (correlation coefficient > 0.8), with weaker or opposite correlations at PreCPB, CPBend, and CPB24h (Figure 5e). Topological overlap matrix (TOM) visualization showed high intramodular connectivity within the lightcyan module (Figure 5f). Strong positive correlation between module membership and gene significance was observed, with highly connected genes also showing strong association with CPB8h (Figure 5g). Patient-level trajectories of the lightcyan module eigengene showed a consistent temporal pattern across individuals, with minimal change at CPBend, a pronounced peak at 8h, and partial decline by 24h (Figure 5h). Integrative Analysis and Machine Learning Select a Three-gene CPB Signature We integrated gene sets derived from single-cell, whole-blood bulk, and neutrophil bulk transcriptomic analyses. GeneSet1 consisted of genes shared between whole-blood bulk RNA-seq (GSE143780) and PBMC single-cell data (GSE262146), while GeneSet2 captured genes overlapping between whole blood and neutrophil bulk RNA-seq (GSE297377). GeneSet3 represented all genes within the CPB-associated lightcyan WGCNA module. GeneSet4 was defined as the intersection between the combined GeneSet1 and GeneSet2 and GeneSet3, representing genes supported by both cross-dataset overlap and co-expression network analysis (Figure 6a). A protein-protein interaction (PPI) network was constructed using the combined GeneSet1 and GeneSet2, and the top 30 hub genes were identified using STRING and Cytoscape (Figure 6b). These hub genes, together with the WGCNA-derived GeneSet4, were subsequently evaluated using three machine-learning models including LASSO, Elastic Net, and Random Forest (Figure 6c). The intersection of features selected by all three models yielded three genes: cluster of differentiation 163 ( CD163 ), interleukin-10 ( IL10 ) and peroxisome proliferator-activated receptor gamma ( PPARG ) (Figure 6d). Density plots showed that this three-gene model effectively separated CPB8h samples from PreCPB/ CPBend samples (Figure 6e). In addition, temporal profiling revealed that all three genes were markedly induced at CPB8h and partially normalized by CPB24h (Figure 6f). Clinical Validation of The Three-gene CPB Signature and Its Association with Postoperative Complications We next measured the expression levels of CD163 , IL10 and PPARG in whole-blood samples at the timepoint of CPB24h. A total of 42 individuals who underwent cardiac surgery necessitating CPB and fulfilled the inclusion criteria were included, and their clinical characteristics and cardiac-lesion types were shown in Table 1 and Table 2, respectively. We assessed whether the signature was associated with postoperative complications, focusing on acute lung injury (ALI). Expression levels of all three genes were significantly higher in patients who developed ALI than in those without ALI (Figure 7a-c). Correlation analyses were performed between gene expression levels and the oxygenation index (OI). All three genes displayed significant correlations with OI (Figure 7d-f, CD163 : r = 0.6523, P ＜ 0.0001; IL10 : r = 0.6397, P ＜ 0.0001; PPARG : r = 0.4935, P = 0.0009). Meanwhile, we found that elevated CD163 (r = 0.5068, P = 0.0006) and IL10 (r = 0.5659, P ＜ 0.0001) level were significantly correlated with increased hospital length of stay (LOS) (Figure 7g-h). PPARG levels also exhibited a positive, though non-significant, correlation with LOS (Figure 7i, r = 0.3010, P = 0.0527). Discussion In this study, we integrated multi-layer transcriptomic datasets [ 8 , 10 , 13 ] with network and machine-learning analyses to delineate early immune responses to CPB and identify a reproducible three-gene signature ( CD163 , IL10 , PPARG ) that captures acute postoperative immune activation. Clinical validation further demonstrated that this signature is robustly induced in pediatric patients undergoing CPB and is associated with postoperative acute lung injury, underscoring its translational relevance. At the systems level, our analyses revealed a coherent pattern in which innate immune and myeloid programs dominate the early postoperative phase, whereas adaptive immunity is consistently attenuated. These findings align with the recognized role of CPB in triggering systemic inflammatory responses and leukocyte activation but extend previous observations by providing cell-type-resolved transcriptomic evidence across multiple datasets. The convergence of whole-blood enrichment patterns with monocyte-macrophage and neutrophil signatures in both single-cell and neutrophil data strongly suggests that myeloid cells constitute the primary drivers of acute inflammatory activation following CPB. These system-level patterns provided the biological rationale for identifying compact and reproducible molecular signatures that capture the core CPB-induced immune response. From a scientific perspective, our study addresses a major limitation of prior transcriptomic analyses, which often examined individual datasets or isolated leukocyte subsets. By integrating orthogonal data modalities, we identify reproducible gene programs that are robust across datasets. WGCNA enables the extraction of a patient-consistent co-expression module that peaked at the CPB8h timepoint, while network and machine-learning analyses further ensure the selected genes are not only differentially expressed but also represent central regulators of the CPB response. Together, these multi-layer analyses converge on a coherent three-gene signature that remains reproducible across datasets, analytical methods, and modeling approaches. On the basis of these integrative and reproducible transcriptomic patterns, we next sought to interpret the biological mechanisms underlying the identified three-gene signature. Previous studies indicated that the systemic inflammatory response syndrome induced by CPB is commonly accompanied by a compensatory anti-inflammatory response. This shift can result in an immunosuppressive state known as immunoparalysis, which leaves patients vulnerable to infections, a leading cause of morbidity after cardiac surgery [ 7 ] [ 34 ]. Neutrophils are typically among the earliest responder cells following CPB and activated neutrophils release proteolytic enzymes, reactive oxygen species, and multiple pro-inflammatory mediators, thereby exacerbating systemic inflammation and contributing to tissue injury [ 4 ] [ 35 ]. Neutrophil-driven innate immune activation after CPB was reported to be related to postoperative organ dysfunction including pulmonary complications[ 5 ], and is increasingly considered as a hallmark of CPB-associated immune dysregulation. At the molecular level, neutrophils display robust activation of innate immune and inflammatory signaling pathways, characterized by upregulation of cytokine production, MAPK signaling, and degranulation-related programs [ 12 ], highlighting their central role in initiating and amplifying postoperative inflammatory responses. Neutrophil-driven inflammation, endothelial activation, and tissue injury signals subsequently provide critical upstream cues for monocyte-macrophage immune responses. In this context, the monocyte-macrophage lineage undergoes a transcriptional reprogramming dominated by immunoregulatory features. CD163, a macrophage-specific hemoglobin scavenger receptor, is markedly upregulated in response to CPB-associated hemolysis and systemic inflammatory stimuli, reflecting macrophage sensing of hemoglobin-heme stress and a shift toward an immunomodulatory phenotype [ 36 , 37 ]. Concurrently, induction of IL-10 represents a negative feedback mechanism activated to restrain excessive inflammation; however, sustained elevation of IL-10 may also contribute to postoperative immunosuppression [ 34 , 38 ]. In addition, activation of PPARG-associated transcriptional programs suggests metabolic-immune coupling within macrophages, promoting alternative macrophage polarization and suppression of pro-inflammatory signaling pathways [ 39 – 41 ]. Taken together, CPB-induced immune responses can be conceptualized as a sequential process characterized by neutrophil-driven inflammatory initiation followed by macrophage-mediated immunoregulatory adaptation. In a subset of patients, this compensatory immunoregulatory program may become exaggerated, resulting in impaired innate immune defense and dysregulated tissue repair. In the present study, the expression levels of CD163 , IL10 , and PPARG were significantly correlated with worsened oxygenation indices and extended hospital length of stay, suggesting that this immune phenotype may contribute to the development of postoperative acute lung injury. This study has several limitations. First, the public datasets used for discovery had relatively small sample sizes, a common limitation in all transcriptomic studies. Nonetheless, the convergence of findings across independent datasets and analytical frameworks supports the robustness of the three-gene signature. Second, although we validated gene expression levels in postoperative clinical samples, we did not assess their paired expression changes before and after CPB, which limits our ability to directly confirm perioperative transcriptional dynamics in our patients. Third, the three-gene signature was identified using transcriptomic profiles at the CPB8h timepoint in public datasets, whereas clinical validation was performed with CPB24h samples due to the availability of clinical specimens. This temporal mismatch may limit direct comparability of transcriptional dynamics across cohorts. Fourth, the clinical validation cohort was modest in size and derived from a single center. The predictive potential of the signature requires evaluation in larger and ideally prospective multicenter cohorts. Lastly, our analysis focused on early postoperative phases; longer-term immune trajectories remain to be explored. Further validation is essential to define the clinical utility of this signature. Conclusion In summary, this multi-layer transcriptomic analysis defines the immune architecture of early CPB responses and identifies a robust three-gene signature with both biological and clinical relevance. These findings highlight the central role of myeloid activation in pediatric CPB and suggest that the three-gene signature may serve as a useful biomarker for early risk stratification in pediatric CPB although subjected to further validation. Declarations Author Contributions: Z.L. and X.W. contributed equally to this work. Z.L. contributed to data curation, formal analysis, visualization, methodology, interpretation of results, and writing of the original draft. X.W. contributed to data curation, validation, visualization, methodology, interpretation of results, and writing of the original draft. X.Z., J.F., L.Y., and H.Y. contributed to validation, data acquisition, and resources. D.M. contributed to methodological guidance and writing—review and editing. L.Yu. and X.L. contributed to writing—review and editing. Q.S. contributed to conceptualization, supervision, project administration, and writing—review and editing. All authors have read and approved the final manuscript and agree to be accountable for their respective contributions. Funding: This work was supported by the Central Guiding Fund for Local Science and Technology Development Projects (No. 2023ZY1058). Institutional Review Board Statement: The study protocol was approved by the Ethics Committee of Children’s Hospital of Zhejiang University (Approval No. 2022-IRB-0054). All procedures were conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from the legal guardians of all participants for clinical patients’ sample analysis part. Data Availability Statement: All transcriptomic datasets analyzed in this study were obtained from the Gene Expression Omnibus (GEO) under accession numbers GSE143780 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE143780), GSE262146 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE262146), and GSE297377 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE297377). No new raw sequencing data were generated in this study, and all analyses were performed using publicly available datasets with standard R packages. No additional custom code or proprietary software was created beyond the scripts described in the Methods section. Acknowledgments: The authors thank all patients and their families for their participation in this study. We also thank the clinical staff of the Department of Pediatric Cardiac Surgery, Children’s Hospital, Zhejiang University School of Medicine, for their assistance with sample collection and clinical data acquisition. Conflicts of Interests: The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. References Backer, C.L.; Overman, D.M.; Dearani, J.A.; Romano, J.C.; Tweddell, J.S.; Kumar, S.R.; Marino, B.S.; Bacha, E.A.; Jaquiss, R.D.B.; Zaidi, A.N.; et al. Recommendations for centers performing pediatric heart surgery in the United States. The Journal of thoracic and cardiovascular surgery 2023 , 166 , 1782-1820, doi:10.1016/j.jtcvs.2023.09.001. Kumar, S.R. et al. The Society of Thoracic Surgeons Congenital Heart Surgery Database: 2022 Update on Outcomes and Research. Ann Thorac Surg 115, 807-819 (2023). 2023 . Liu, H.; Hu, Y.J.; Zheng, S.Q.; Chen, T.; Zeng, Z.H.; Wu, D.D.; Zhao, S.; Zeng, B.; Liu, Z.G.; Shao, Y.F. Effect of perfusate oxygenation on inflammatory response in congenital heart disease children from low versus high altitude. The Journal of thoracic and cardiovascular surgery 2021 , 161 , 2180-2190, doi:10.1016/j.jtcvs.2020.05.108. Esper, S.A.; Subramaniam, K.; Tanaka, K.A. Pathophysiology of Cardiopulmonary Bypass: Current Strategies for the Prevention and Treatment of Anemia, Coagulopathy, and Organ Dysfunction. Semin Cardiothorac Vasc Anesth 2014 , 18 , 161-176, doi:10.1177/1089253214532375. Mahle, W.T.; Matthews, E.; Kanter, K.R.; Kogon, B.E.; Hamrick, S.E.; Strickland, M.J. Inflammatory response after neonatal cardiac surgery and its relationship to clinical outcomes. Ann Thorac Surg 2014 , 97 , 950-956, doi:10.1016/j.athoracsur.2013.10.069. Appachi; E; Mossad; E; Mee; R.B. & Bokesch, P. Perioperative serum interleukins in neonates with hypoplastic left-heart syndrome and transposition of the great arteries. J Cardiothorac Vasc Anesth 21, 184-190 (2007). 2007 . Lesouhaitier, M.; Belicard, F.; Tadié, J.M. Cardiopulmonary bypass and VA-ECMO induced immune dysfunction: common features and differences, a narrative review. Critical care (London, England) 2024 , 28 , 300, doi:10.1186/s13054-024-05058-z. Tu, L.N.; Hsieh, L.; Kajimoto, M.; Charette, K.; Kibiryeva, N.; Forero, A.; Hampson, S.; Marshall, J.A.; O'Brien, J.; Scatena, M.; et al. Shear stress associated with cardiopulmonary bypass induces expression of inflammatory cytokines and necroptosis in monocytes. JCI insight 2021 , 6 , doi:10.1172/jci.insight.141341. Jain, P.N.; Robertson, M.; Lasa, J.J.; Shekerdemian, L.; Guffey, D.; Zhang, Y.; Lingappan, K.; Checchia, P.; Coarfa, C. Altered metabolic and inflammatory transcriptomics after cardiac surgery in neonates with congenital heart disease. Sci Rep 2021 , 11 , 4965, doi:10.1038/s41598-021-83882-x. Li, W.; Tu, L.N.; Hsieh, L.; Smith, J.R.; Yeh, Y.-T.; Sinyagin, A.; Ghassemian, M.; Timms, A.; Charette, K.; Mauchley, D.; et al. Cardiopulmonary bypass activates classical monocytes via shear-mediated activation of Store-Operated Calcium Entry. bioRxiv 2024 , 2022.2005.2004.490549, doi:10.1101/2022.05.04.490549. Chen, H.; Liu, J.; Wu, Y.; Jiang, L.; Tang, M.; Wang, X.; Fang, X.; Wang, X. Weighted gene co-expression identification of CDKN1A as a hub inflammation gene following cardiopulmonary bypass in children with congenital heart disease. Front Surg 2022 , 9 , 963850, doi:10.3389/fsurg.2022.963850. Maisat W, H.L.S.S.S.Y.C.K.S.V.P.H. Neutrophil extracellular traps formation is associated with postoperative complications in congenital cardiac surgery. Pediatr Res. (2024):1–14. doi:10.1038/s41390-024-03717-z 2024 . Maisat, W.; Sandhu, S.; Kim, S.; Van Pelt, H.; Kong, S.W.; Ibla, J.; Yuki, K. Neutrophil Myo5c gene downregulation is associated with postoperative organ dysfunction following pediatric cardiac surgery with cardiopulmonary bypass. Front Cardiovasc Med 2025 , 12 , 1380606, doi:10.3389/fcvm.2025.1380606. Raggi, F.; Cangelosi, D.; Becherini, P.; Blengio, F.; Morini, M.; Acquaviva, M.; Belli, M.L.; Panizzon, G.; Cervo, G.; Varesio, L.; et al. Transcriptome analysis defines myocardium gene signatures in children with ToF and ASD and reveals disease-specific molecular reprogramming in response to surgery with cardiopulmonary bypass. J Transl Med 2020 , 18 , 21, doi:10.1186/s12967-020-02210-5. Wang, Y.; Chen, L.; Yao, C.; Wang, T.; Wu, J.; Shang, Y.; Li, B.; Xia, H.; Huang, S.; Wang, F.; et al. Early plasma proteomic biomarkers and prediction model of acute respiratory distress syndrome after cardiopulmonary bypass: a prospective nested cohort study. International journal of surgery (London, England) 2023 , 109 , 2561-2573, doi:10.1097/js9.0000000000000434. Love, M.I.; Huber, W.; Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biology 2014 , 15 , 550-550. Wickham, H. ggplot2: Elegant Graphics for Data Analysis. Springer 2016 . Gu, Z.; Eils, R.; Schlesner, M. Complex heatmaps reveal patterns and correlations in multidimensional genomic data. Bioinformatics 2016 , 32 , 2847-2849. Stuart, T.; Butler, A.; Hoffman, P.; et al. Comprehensive Integration of Single-Cell Data. Cell 2019 , 177 , 1888-1902. Amezquita, R.A.; Lun, A.T.L.; Becht, E.; Carey, V.J.; Carpp, L.N.; Geistlinger, L.; Marini, F.; Rue-Albrecht, K.; Risso, D.; Soneson, C.; et al. Orchestrating single-cell analysis with Bioconductor. Nat Methods 2020 , 17 , 137-145, doi:10.1038/s41592-019-0654-x. Xu, S.; Hu, E.; Cai, Y.; Xie, Z.; Luo, X.; Zhan, L.; Tang, W.; Wang, Q.; Liu, B.; Wang, R.; et al. Using clusterProfiler to characterize multiomics data. Nat Protoc 2024 , 19 , 3292-3320, doi:10.1038/s41596-024-01020-z. Liberzon, A.; Birger, C.; Thorvaldsdottir, H.; Ghandi, M.; Mesirov, J.P.; Tamayo, P. The Molecular Signatures Database (MSigDB) hallmark gene set collection. Cell Syst 2015 , 1 , 417-425, doi:10.1016/j.cels.2015.12.004. Yu, G.; He, Q.Y. ReactomePA: an R/Bioconductor package for reactome pathway analysis and visualization. Molecular BioSystems 2016 , 12 , 477-479. Kanehisa, M.; Furumichi, M.; Sato, Y.; Kawashima, M.; Ishiguro-Watanabe, M. KEGG for taxonomy-based analysis of pathways and genomes. Nucleic Acids Res 2023 , 51 , D587-d592, doi:10.1093/nar/gkac963. Langfelder, P.; Horvath, S. WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics 2008 , 9 , 559-559. Szklarczyk, D.; et al. The STRING database in 2021: customizable protein–protein networks and functional characterization. Nucleic Acids Research 2021 , 49 , D605-D612. Shannon, P.; et al. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Research 2003 , 13 , 2498-2504. Chin, C.H.; et al. cytoHubba: identifying hub objects and sub-networks from complex interactome. BMC Systems Biology 2014 , 8 , S11-S11. Friedman, J.; Hastie, T.; Tibshirani, R. Regularization Paths for Generalized Linear Models via Coordinate Descent. Journal of Statistical Software 2010 , 33 , 1-22. Breiman, L. Random forests. Machine Learning 2001 , 45 , 5-32. Emeriaud, G.; López-Fernández, Y.M.; Iyer, N.P.; Bembea, M.M.; Agulnik, A.; Barbaro, R.P.; Baudin, F.; Bhalla, A.; Brunow de Carvalho, W.; Carroll, C.L.; et al. Executive Summary of the Second International Guidelines for the Diagnosis and Management of Pediatric Acute Respiratory Distress Syndrome (PALICC-2). Pediatr Crit Care Med 2023 , 24 , 143-168, doi:10.1097/pcc.0000000000003147. Ad, N.; Snir, E.; Katz, J.; Birk, E.; Vidne, B.A. Use of the modified technique of ultrafiltration in pediatric open-heart surgery: a prospective study. Isr J Med Sci 1996 , 32 , 1326-1331. Chew, M.S.; Brix-Christensen, V.; Ravn, H.B.; Brandslund, I.; Ditlevsen, E.; Pedersen, J.; Hjortholm, K.; Hansen, O.K.; Tonnesen, E.; Hjortdal, V.E. Effect of modified ultrafiltration on the inflammatory response in paediatric open-heart surgery: a prospective, randomized study. Perfusion 2002 , 17 , 327-333, doi:10.1191/0267659102pf595oa. Liu, C.H.; Huang, Z.H.; Huang, S.C.; Jou, T.S. Endocytosis of peroxiredoxin 1 links sterile inflammation to immunoparalysis in pediatric patients following cardiopulmonary bypass. Redox biology 2021 , 46 , 102086, doi:10.1016/j.redox.2021.102086. Chew, M.S.; Brandslund, I.; Brix-Christensen, V.; Ravn, H.B.; Hjortdal, V.E.; Pedersen, J.; Hjortdal, K.; Hansen, O.K.; Tonnesen, E. Tissue injury and the inflammatory response to pediatric cardiac surgery with cardiopulmonary bypass: a descriptive study. Anesthesiology 2001 , 94 , 745-753; discussion 745A, doi:10.1097/00000542-200105000-00010. Etzerodt, A.; Moestrup, S.K. CD163: a scavenger receptor clearing hemoglobin–haptoglobin complexes. Immunology 2013 . Philippidis, P.; Mason, J.C.; Evans, B.J.; Nadra, I.; Taylor, K.M.; Haskard, D.O.; Landis, R.C. Hemoglobin scavenger receptor CD163 mediates interleukin-10 release and heme oxygenase-1 synthesis: antiinflammatory monocyte-macrophage responses in vitro, in resolving skin blisters in vivo, and after cardiopulmonary bypass surgery. Circulation research 2004 , 94 , 119-126, doi:10.1161/01.Res.0000109414.78907.F9. Moore, K.W.; et al. IL-10 and the regulation of immune responses. Annual Review of Immunology 2001 . Odegaard, J.I.; Ricardo-Gonzalez, R.R.; Goforth, M.H.; Morel, C.R.; Subramanian, V.; Mukundan, L.; Red Eagle, A.; Vats, D.; Brombacher, F.; Ferrante, A.W.; et al. Macrophage-specific PPARgamma controls alternative activation and improves insulin resistance. Nature 2007 , 447 , 1116-1120, doi:10.1038/nature05894. Bouhlel, M.A.; Derudas, B.; Rigamonti, E.; Dièvart, R.; Brozek, J.; Haulon, S.; Zawadzki, C.; Jude, B.; Torpier, G.; Marx, N.; et al. PPARgamma activation primes human monocytes into alternative M2 macrophages with anti-inflammatory properties. Cell Metab 2007 , 6 , 137-143, doi:10.1016/j.cmet.2007.06.010. Liu, B.; Liang, G.; Xu, G.; Liu, D.; Cai, Q.; Gao, Z. Intervention of rosiglitazone on myocardium Glut-4 mRNA expression during ischemia-reperfusion injury in cardio-pulmonary bypass in dogs. Molecular and cellular biochemistry 2013 , 373 , 279-284, doi:10.1007/s11010-012-1501-x. Tables Table 1 ． Demographic and clinical characteristics Study cohort (n=42) ALI (n=18) Non-ALI (n=24) P value Sex (male; n [%]) 27(64.3%) 11(61.1%) 16(66.7%) 0.754 Age (months) 8.69±1.626 4.83±0.933 11.58±2.63 0.038 Weight (kg) 7.2±0.451 6.483±0.448 7.375±0.709 0.193 Operation time (minutes) 155.5±6.142 171.1±11.32 143.7±5.716 0.007 CPB time(minutes) 88.93±5.359 98.72±10.68 81.58±4.57 0.117 AC time (minutes) 55.24±4.543 63.67±9.147 48.92±3.736 0.109 Hospital LOS (days) 16.79±2.180 18.61±1.684 15.42±3.614 0.0021 Data are presented as number of patients (%), mean ± SEM, or counts, as appropriate. ALI, acute lung injury; CPB, cardiopulmonary bypass; LOS, length of stay; AC, aortic cross-clamp. Table 2 ． Cardiac lesion types Study cohort ALI Non-ALI VSD 20 10(50%) 10(50%) ASD 7 1(14%) 6(86) TOF 9 5(56%) 4(44%) ASD plus VSD 5 1(20%) 4(80%) PS 1 1(100%) 0(1%) Data are presented as counts (%). ASD, atrial septal defect; PS, pulmonary stenosis; TOF, tetralogy of Fallot; VSD, ventricular septal defect. Additional Declarations No competing interests reported. Supplementary Files SupplementaryMaterials.docx Cite Share Download PDF Status: Under Revision Version 1 posted Editorial decision: Revision requested 05 May, 2026 Reviews received at journal 04 May, 2026 Reviewers agreed at journal 29 Apr, 2026 Reviewers agreed at journal 28 Apr, 2026 Reviewers agreed at journal 28 Apr, 2026 Reviews received at journal 26 Feb, 2026 Reviews received at journal 21 Feb, 2026 Reviewers agreed at journal 14 Feb, 2026 Reviewers agreed at journal 09 Feb, 2026 Reviewers invited by journal 06 Feb, 2026 Editor assigned by journal 06 Feb, 2026 Editor invited by journal 06 Feb, 2026 Submission checks completed at journal 06 Feb, 2026 First submitted to journal 06 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-8751380","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":588565354,"identity":"249e6c29-9d1f-4100-8c84-63c835367ac3","order_by":0,"name":"Zhicong Liu","email":"","orcid":"","institution":"Children's Hospital of Zhejiang University","correspondingAuthor":false,"prefix":"","firstName":"Zhicong","middleName":"","lastName":"Liu","suffix":""},{"id":588565355,"identity":"e227e137-1a72-49df-8889-7e821294f084","order_by":1,"name":"Xueke Wang","email":"","orcid":"","institution":"Children's Hospital of Zhejiang University","correspondingAuthor":false,"prefix":"","firstName":"Xueke","middleName":"","lastName":"Wang","suffix":""},{"id":588565358,"identity":"5b025335-88fe-4090-b79d-8f6a75621c2c","order_by":2,"name":"Xiaohui Zhong","email":"","orcid":"","institution":"Children's Hospital of Zhejiang University","correspondingAuthor":false,"prefix":"","firstName":"Xiaohui","middleName":"","lastName":"Zhong","suffix":""},{"id":588565360,"identity":"9d171c94-def8-4d3a-a475-0c70a72c7431","order_by":3,"name":"Jiajie Fan","email":"","orcid":"","institution":"Children's Hospital of Zhejiang University","correspondingAuthor":false,"prefix":"","firstName":"Jiajie","middleName":"","lastName":"Fan","suffix":""},{"id":588565365,"identity":"a7f1c80c-2648-4cd8-b266-234bbcbd186e","order_by":4,"name":"Liyang Ying","email":"","orcid":"","institution":"Children's Hospital of Zhejiang University","correspondingAuthor":false,"prefix":"","firstName":"Liyang","middleName":"","lastName":"Ying","suffix":""},{"id":588565367,"identity":"a39184e3-4571-44d9-9b61-0c419cf9116d","order_by":5,"name":"Hui Ye","email":"","orcid":"","institution":"Children's Hospital of Zhejiang University","correspondingAuthor":false,"prefix":"","firstName":"Hui","middleName":"","lastName":"Ye","suffix":""},{"id":588565368,"identity":"f75693b3-d688-4f12-9a0d-69fe688cadb2","order_by":6,"name":"Daqing Ma","email":"","orcid":"","institution":"Children's Hospital of Zhejiang University","correspondingAuthor":false,"prefix":"","firstName":"Daqing","middleName":"","lastName":"Ma","suffix":""},{"id":588565369,"identity":"bfb7b2b6-10b3-4a38-ac89-2bff551d3f7c","order_by":7,"name":"Lan Yu","email":"","orcid":"","institution":"Children's Hospital of Zhejiang University","correspondingAuthor":false,"prefix":"","firstName":"Lan","middleName":"","lastName":"Yu","suffix":""},{"id":588565371,"identity":"6ee4373d-21c2-43a4-b76b-d48b8c7c9e3e","order_by":8,"name":"Xiwang Liu","email":"","orcid":"","institution":"Children's Hospital of Zhejiang University","correspondingAuthor":false,"prefix":"","firstName":"Xiwang","middleName":"","lastName":"Liu","suffix":""},{"id":588565373,"identity":"52493181-8ada-47a3-ae93-decc39505b30","order_by":9,"name":"Qiang Shu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA3UlEQVRIiWNgGAWjYDACCSBOAGJ+CMVMghbJBpK0gIDBATBFhBb52T2GNx7U3LHbfP7AMwmGCuvEBvazB/BqMbhzxtgi4diz5G0HDqRJMJxJT2zgyUvAr0Uix0wige1wstnBhjQJxrbDiQ0SPAb4HTYDpOXf4WTjZgagln9EaGG4AdSS2HbYzoANpKWBCC0GN9KKLRL7DidInGFIBnoq3biNJ4eQw5I33vzx7bA9f/+ZxBsfaqxl+9nPEHAYAyRqEhsYeBLAkclGUD1Uiz0DA/sBYhSPglEwCkbBCAQADjhFwpVLQSUAAAAASUVORK5CYII=","orcid":"","institution":"Children's Hospital of Zhejiang University","correspondingAuthor":true,"prefix":"","firstName":"Qiang","middleName":"","lastName":"Shu","suffix":""}],"badges":[],"createdAt":"2026-01-31 16:39:07","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8751380/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8751380/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":102439960,"identity":"c73695de-0da2-45a7-8542-3b4b0988ac73","added_by":"auto","created_at":"2026-02-11 16:44:37","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":36069,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eOverview of study design and analytical workflow.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(a) Bioinformatics integration of DEGs, time-course clusters, and WGCNA modules and PPI networks to identify CPB-related core genes. Intersected genes were evaluated using machine-learning models, followed by clinical validation in pediatric CPB samples. DEG, differentially expressed gene; WGCNA, weighted gene co-expression network analysis; PPI, protein-protein interaction network.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8751380/v1/d0e388c504c2544a0dec5147.png"},{"id":102439966,"identity":"c4554db7-b196-468c-946d-d4f7199b125f","added_by":"auto","created_at":"2026-02-11 16:44:38","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":217841,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMulti-dataset transcriptomic responses to CPB.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(a) Sampling scheme of the three public datasets, including whole-blood bulk RNA-seq (GSE143780), PBMC scRNA-seq (GSE262146), and neutrophil bulk RNA-seq (GSE297377). (b) Time-course gene expression clusters from GSE143780 showing representative temporal patterns across pre-CPB, CPBend, 8h, and 24h.\u003c/p\u003e\n\u003cp\u003e(c-e) Heatmaps of differentially expressed genes comparing post-CPB with pre-CPB samples in (c) GSE143780 (CPB 8h vs Pre), (d) GSE262146 (8h vs Pre), and (e) GSE297377 (Post vs Pre).\u003c/p\u003e\n\u003cp\u003e(f-h) Volcano plots showing significantly upregulated (red) and downregulated (blue) genes for the corresponding comparisons in (c-e). Each dot represents one gene; thresholds are based on adjusted \u003cem\u003eP\u003c/em\u003e values and fold changes.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8751380/v1/33391b533b627e2eb514b0e8.png"},{"id":102439962,"identity":"55d11912-604c-4ce7-b17e-7302303ee3f3","added_by":"auto","created_at":"2026-02-11 16:44:37","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":122739,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePathway-level transcriptomic responses to CPB in whole-blood leukocytes.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(a, b) GO biological process (GOBP) enrichment of upregulated (a) and downregulated (b) genes at different CPB timepoints (CPBend, CPB8h, CPB24h) compared with pre-CPB samples in GSE143780. Dot size represents gene counts and color indicates statistical significance.\u003c/p\u003e\n\u003cp\u003e(c, d) KEGG pathway enrichment of upregulated (c) and downregulated (d) genes across the same comparisons. KEGG pathway enrichment was performed using the KEGG database; no KEGG pathway maps were reproduced.\u003c/p\u003e\n\u003cp\u003e(e, f) Gene set enrichment analysis (GSEA) of whole-blood leukocyte transcriptomes comparing CPB 8h versus pre-CPB, showing selected upregulated (e) and downregulated (f) pathways.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8751380/v1/9dce0e4c1290f0f25b17ab3f.png"},{"id":102439964,"identity":"46d2bfec-43f8-4b04-b461-4dc9bb85eee5","added_by":"auto","created_at":"2026-02-11 16:44:37","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":179166,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eImmune lineage-level transcriptomic responses to CPB across single-cell and neutrophil datasets.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(a) UMAP visualization of PBMCs from GSE262146 at preCPB, postCPB, 8h, and 24h, showing major immune lineages including T lymphocytes, B lymphocytes, NK cells, monocyte-macrophage lineage, dendritic cells, progenitors, and cycling cells.\u003c/p\u003e\n\u003cp\u003e(b) Relative proportions of each immune lineage across the four timepoints.\u003c/p\u003e\n\u003cp\u003e(c, d) Numbers of upregulated (c) and downregulated (d) DEGs in each immune lineage at postCPB, 8h, and 24h compared with preCPB.\u003c/p\u003e\n\u003cp\u003e(e) GOBP enrichment of upregulated and downregulated DEGs in the monocyte-macrophage lineage at postCPB, 8h, and 24h versus pre-CPB.\u003c/p\u003e\n\u003cp\u003e(f) GOBP enrichment of DEGs in the neutrophil lineage comparing postCPB with preCPB in GSE297377.\u003c/p\u003e\n\u003cp\u003e(g) KEGG pathway enrichment of neutrophil-lineage DEGs in GSE297377 comparing postCPB with preCPB. KEGG pathway enrichment was performed using the KEGG database; no KEGG pathway maps were reproduced.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8751380/v1/3d4862e0aa299cee7f430396.png"},{"id":103056356,"identity":"eb7c6eb1-c060-4e12-894a-3fb92e2095fa","added_by":"auto","created_at":"2026-02-20 09:07:43","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":148153,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eWGCNA identifies CPB-associated co-expression modules in whole-blood transcriptomes.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(a) Sample clustering dendrogram of whole-blood transcriptomes in GSE143780, showing overall sample similarity across preCPB, CPBend, 8h, and 24h timepoints.\u003c/p\u003e\n\u003cp\u003e(b) Selection of the soft-thresholding power based on scale-free topology fit index (left) and mean connectivity (right).\u003c/p\u003e\n\u003cp\u003e(c) Clustering dendrogram of module eigengenes showing relationships among identified co-expression modules.\u003c/p\u003e\n\u003cp\u003e(d) Heatmap of pairwise correlations among module eigengenes, indicating modules with similar expression patterns.\u003c/p\u003e\n\u003cp\u003e(e) Module-trait heatmap showing correlations between module eigengenes and clinical timepoints (preCPB, CPBend, 8h, 24h). Numbers indicate correlation coefficients and P values.\u003c/p\u003e\n\u003cp\u003e(f) Topological overlap matrix (TOM) heatmap of genes within the CPB associated lightcyan module.\u003c/p\u003e\n\u003cp\u003e(g) Scatter plot showing the relationship between module membership (MM) in the lightcyan module and gene significance (GS) for CPB 8h versus preCPB.\u003c/p\u003e\n\u003cp\u003e(h) Temporal dynamics of the lightcyan module eigengene across preCPB, CPBend, 8h, and 24h in individual patients.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8751380/v1/5635e85aedc23386019c8e76.png"},{"id":102439967,"identity":"796814ee-1ec7-4188-a5a8-457f1c4dc84d","added_by":"auto","created_at":"2026-02-11 16:44:38","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":196490,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eIntegrated multi-dataset analysis identifies a 3-gene CPB signature.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(a) Venn diagrams showing the overlapping gene sets derived from GSE262146, GSE143780, GSE297377, and WGCNA modules. Four integrated gene sets (GeneSet1-4) were obtained based on multi-dataset intersections.\u003c/p\u003e\n\u003cp\u003e(b) Protein-protein interaction (PPI) network of the top 30 hub genes derived from the integrated gene set. Node color intensity reflects network centrality.\u003c/p\u003e\n\u003cp\u003e(c) Heatmap showing gene importance scores across three machine-learning models (LASSO, Elastic Net, and Random Forest).\u003c/p\u003e\n\u003cp\u003e(d) Venn diagram illustrating the overlap of genes selected by the three machine-learning models, yielding a 3-gene signature (\u003cem\u003eCD163\u003c/em\u003e, \u003cem\u003eIL10\u003c/em\u003e, \u003cem\u003ePPARG\u003c/em\u003e).\u003c/p\u003e\n\u003cp\u003e(e) Predicted probabilities based on the 3-gene signature, showing separation between preCPB/ CPBend samples and CPB8h samples.\u003c/p\u003e\n\u003cp\u003e(f) Temporal expression dynamics of the three signature genes (\u003cem\u003eCD163\u003c/em\u003e, \u003cem\u003eIL10\u003c/em\u003e, \u003cem\u003ePPARG\u003c/em\u003e) across preCPB, CPBend, 8h, and 24h in GSE143780.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-8751380/v1/2f17f26694083a521495dc62.png"},{"id":102746004,"identity":"764cd729-0ac9-43fa-9bab-bb7c67a6b1bc","added_by":"auto","created_at":"2026-02-16 08:55:09","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":58103,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eClinical validation of the 3-gene signature and correlations with postoperative outcomes.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(a-c) Gene expression level of \u003cem\u003eCD163, IL10 and PPARG\u003c/em\u003e in 18 patients with ALI and 24 patients without ALI. Each dot represents one individual. Data were analyzed by the Student’s t test or Mann-Whitney U test, as appropriate.\u003c/p\u003e\n\u003cp\u003e(d-f) Correlation of \u003cem\u003eCD163, IL10 and PPARG\u003c/em\u003e mRNA level in patients with oxygenation index (n = 42). Each dot represents one individual. Data were analyzed by Pearson or Spearman correlation test.\u003c/p\u003e\n\u003cp\u003e(g-i) Correlation of \u003cem\u003eCD163, IL10 and PPARG\u003c/em\u003e mRNA level in patients with hospital LOS (n = 42). Each dot represents one individual. Data were analyzed by Pearson or Spearman correlation test.\u003c/p\u003e","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-8751380/v1/23ada8cdca51b532f4d50612.png"},{"id":103056668,"identity":"d753a3fa-5046-4139-87dd-26a376513b5b","added_by":"auto","created_at":"2026-02-20 09:23:59","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2280873,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8751380/v1/7483484a-91c8-4958-8ce0-5dda85a09d96.pdf"},{"id":102439961,"identity":"ee0a1dda-5bd2-4202-bc3b-ea0bd12ebba0","added_by":"auto","created_at":"2026-02-11 16:44:37","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":838480,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterials.docx","url":"https://assets-eu.researchsquare.com/files/rs-8751380/v1/e49dee0aac125e6634101cde.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Identification of biomarkers associated with acute lung injury after cardiopulmonary bypass by integrative transcriptomic analysis and clinical validation","fulltext":[{"header":"Introduction","content":"\u003cp\u003eCardiopulmonary bypass (CPB) is widely used in pediatric cardiac surgery including the correction of congenital heart defect [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. However, blood exposure to artificial surfaces, ischemia-reperfusion injury, and hemodilution trigger systemic inflammatory and immune responses that are closely associated with postoperative complications [\u003cspan additionalcitationids=\"CR4\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. These processes involve innate immune activation, cytokine release, leukocyte recruitment, and endothelial dysfunction [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e], yet the molecular mechanisms underlying these alterations remain elusive.\u003c/p\u003e \u003cp\u003eAlthough several transcriptomic studies reported CPB-associated changes in whole blood [\u003cspan additionalcitationids=\"CR9\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e], leukocyte subsets [\u003cspan additionalcitationids=\"CR12\" citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e], or pediatric myocardial tissue [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e], the available datasets remain sparse, heterogeneous, and typically limited to single cohorts. As a result, the cell-type-resolved and cross-dataset transcriptional architecture of early postoperative immune responses in children remains unknown. Systems-level investigations integrating bulk, single-cell, and compartment-specific transcriptomic layers are particularly scarce. Current clinical biomarkers primarily reflect late inflammatory changes, underscoring the need for early and robust molecular signatures that capture core CPB-induced transcriptional alterations [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eTo address this gap, we integrated bulk and single-cell transcriptomic datasets from multiple immune compartments to delineate early CPB-induced immune responses and identify reproducible molecular signatures. We further validated these transcriptomic findings in paired pediatric clinical samples and examined their association with postoperative complications and clinical outcomes.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eTranscriptomic Datasets\u003c/h2\u003e \u003cp\u003eThree publicly available transcriptomic datasets profiling human peripheral blood responses to cardiopulmonary bypass (CPB) were analyzed in this study, including whole-blood bulk RNA-seq (GSE143780) [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e], PBMC single-cell RNA-seq (GSE262146) [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e], and neutrophil bulk RNA-seq (GSE297377) [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. All raw or processed matrices were obtained directly from the Gene Expression Omnibus (GEO). GSE143780 and GSE297377 were downloaded as count matrices, while GSE262146 was processed from the \u003cem\u003efiltered_feature_bc_matrix.h5\u003c/em\u003e file. For consistency across datasets, the single-cell data were re-processed without reference with the original authors\u0026rsquo; annotation. These datasets were selected because they captured comparable perioperative sampling windows and enabled cross-compartment examination of acute transcriptional responses following CPB.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eBulk RNA-seq Processing and Differential Expression Analysis\u003c/h3\u003e\n\u003cp\u003eBulk RNA-seq analyses for GSE143780 and GSE297377 were performed in R (version 4.5.2). Gene-level count matrices were normalized and variance-stabilized with the DESeq2 package [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e] (version 1.48.2). Differential expression analysis was conducted using the Wald test implemented in DESeq2, and genes with |log2 fold change| \u0026ge; 1 and adjusted p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 were considered significant. All visualization procedures for the bulk RNA-seq datasets including volcano plots and heatmaps were performed using ggplot2 [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e] (version 4.0.0) and ComplexHeatmap [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e] (version 2.24.0).\u003c/p\u003e\n\u003ch3\u003eSingle-cell RNA-seq Processing\u003c/h3\u003e\n\u003cp\u003eSingle-cell data from GSE262146 were processed using Seurat [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e] (version 5.3.1) and SingleCellExperiment [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e] (version 1.30.1). Standard quality-control thresholds were applied to remove cells with low gene counts, excessive mitochondrial percentages, or extreme UMI levels. Normalization, scaling, PCA, clustering, and UMAP embeddings followed the Seurat workflow using the first 20 principal components. Lineage-specific enrichment analyses were conducted using clusterProfiler [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e] (version 4.16.0) and msigdbr [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e] (version 25.1.1).\u003c/p\u003e\n\u003ch3\u003eFunctional Enrichment Analyses\u003c/h3\u003e\n\u003cp\u003eFunctional enrichment analyses were conducted with clusterProfiler, ReactomePA [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e] (version 1.52.0), and msigdbr. GO Biological Process and KEGG pathway overrepresentation analyses used DEGs identified from the Wald test. KEGG pathway enrichment analysis was performed using the KEGG database [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. Gene Set Enrichment Analysis (GSEA) utilized the full ranked gene list with default parameters (minGSSize\u0026thinsp;=\u0026thinsp;10, maxGSSize\u0026thinsp;=\u0026thinsp;500, and 1000 permutations). Because CPB8h exhibited the strongest early transcriptional changes, additional enrichment analyses were performed specifically for this comparison to assess phase-specific pathway activation.\u003c/p\u003e\n\u003ch3\u003eWeighted Gene Co-expression Network Analysis\u003c/h3\u003e\n\u003cp\u003eWeighted gene co-expression network analysis was conducted on the log2-transformed expression matrix of GSE143780 using the WGCNA package [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e] (version 1.73). A signed hybrid network was constructed after removing low-variance genes. The soft-thresholding power was set to 9 based on scale-free topology criteria. Modules were constructed using blockwiseModules with minModuleSize\u0026thinsp;=\u0026thinsp;30, mergeCutHeight\u0026thinsp;=\u0026thinsp;0.25, and default settings otherwise. Module eigengenes were correlated with perioperative CPB phases (PreCPB, CPBend, CPB8h, and CPB24h), and the module showing the strongest association with CPB8h was designated as the CPB-responsive module for downstream analyses.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eProtein-protein Interaction (PPI) Network Analysis\u003c/h2\u003e \u003cp\u003eGenes shared across bulk and single-cell DEG intersections were analyzed using STRING [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e] (version 12.0) with default confidence-score thresholds. The interaction network was imported into Cytoscape [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e] (version 3.10.3), and hub genes were ranked using the cytoHubba plugin [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e] \u003cem\u003evia\u003c/em\u003e the Maximal Clique Centrality (MCC) algorithm. The top 30 MCC-ranked genes were retained as putative regulatory hubs. All network visualizations were produced directly within Cytoscape using the \u0026ldquo;circular layout\u0026rdquo;.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eMachine-learning-based Gene Selection\u003c/h3\u003e\n\u003cp\u003eMachine-learning models were trained using a feature set derived from two sources. First, the top 30 hub genes ranked by the Maximal Clique Centrality (MCC) algorithm in the cytoHubba plugin were extracted from the STRING-derived PPI network. Second, genes belonging to the CPB-responsive WGCNA module from GSE143780 were intersected with genes jointly identified across whole-blood (GSE143780), neutrophil (GSE297377), and PBMC single-cell (GSE262146) differential expression analyses. This intersection set (GeneSet4) represented genes supported by both co-expression structure and multi-compartment transcriptomic reproducibility. The union of the PPI top 30 genes and GeneSet4 formed the feature matrix used for model training.\u003c/p\u003e \u003cp\u003eTwo regularized logistic regression models\u0026mdash;LASSO (α\u0026thinsp;=\u0026thinsp;1.0) and Elastic Net (0\u0026thinsp;\u0026lt;\u0026thinsp;α\u0026thinsp;\u0026lt;\u0026thinsp;1.0)\u0026mdash;were implemented with the glmnet package [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e] (version 4.1.10) with optimal λ selected by 10-fold cross-validation. In parallel, a Random Forest classifier was trained using the randomForest package [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e] (version 4.7.1.2) with 1000 trees, and gene importance was quantified using MeanDecreaseGini metrics. Genes consistently prioritized across all three models were considered robust predictors of acute CPB activation and were consolidated into the final three-gene signature.\u003c/p\u003e\n\u003ch3\u003eCollecting Patients’ Samples\u003c/h3\u003e\n\u003cp\u003eAfter obtained the approval by the Ethics Committee of Children\u0026rsquo;s Hospital (Approval No. 2022-IRB-0054), Zhejiang University School of Medicine, Zhejiang, China and written informed consent from their parents, a total of 42 patients under 3 years old who underwent cardiac surgery under CPB were enrolled. Their blood samples were collected into ethylene diamine tetraacetic acid vacuum tubes 24 hours after CPB. The exclusion criteria were as follows: premature infants, those with liver and kidney dysfunction, those with major chromosome abnormalities, those with pulmonary inflammation before surgery, those who required extracorporeal membrane oxygenation support after surgery, those who died because of cardiac dysfunction, or those who refused to provide informed consent.\u003c/p\u003e \u003cp\u003eThe Second International Guidelines for the Diagnosis and Management of Pediatric Acute Respiratory Distress Syndrome (PALICC-2) [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e] were used to categorize the patients into the ALI group and non-ALI group. In brief, this criterion includes new opacities consistent with acute pulmonary parenchymal disease which are not due primarily to atelectasis or pleural effusion seen on a chest radiograph, an oxygenation index (OI)\u0026thinsp;\u0026ge;\u0026thinsp;4, and the absence of cardiogenic pulmonary edema. The formula for OI: mean airway pressure (MAP) (cm H\u003csub\u003e2\u003c/sub\u003eO) \u0026times; FiO\u003csub\u003e2\u003c/sub\u003e/PaO\u003csub\u003e2\u003c/sub\u003e (mm Hg).\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eQuantitative Real-time Polymerase Chain Reaction (qRT-PCR)\u003c/h2\u003e \u003cp\u003eGenomic DNA was extracted from whole blood samples from the above patient cohort with a commercially available DNA isolation kit (QIAGEN). The obtained cDNA was amplified using TB Green Premix Ex Taq (Takara) on a Roche LightCycler480 PCR System. The expression levels of \u003cem\u003eCD163\u003c/em\u003e, \u003cem\u003eIL10\u003c/em\u003e and \u003cem\u003ePPARG\u003c/em\u003e transcripts were normalized to the expression level of ACTB. The gene relative expression was calculated using the 2\u003csup\u003e\u0026minus;ΔΔCt\u003c/sup\u003e method. Primer sequences for these genes are listed in Supplementary Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eStatistical Analysis\u003c/h2\u003e \u003cp\u003eContinuous data were tested for normal distribution with the one-sample Kolmogorov-Smirnov test. Variables were presented as mean values and standard deviations if normally distributed, and otherwise, as median values (interquartile range). The Student\u0026rsquo;s t test or Mann-Whitney U test were used to determine the significance of variable differences between the two groups where appropriate. A Pearson or Spearman correlation test was performed to determine the correlation between biomarker data and clinical outcomes. For all analyses, a \u003cem\u003eP\u003c/em\u003e value less than 0.05 was considered to be statistically significant.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eTimepoint Harmonization across Datasets\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe analyzed three complementary transcriptomic datasets, including whole-blood bulk RNA-seq (GSE143780) [8], PBMC single-cell RNA-seq (GSE262146) [10], and neutrophil bulk RNA-seq (GSE297377) [13]. An overview of the study design and analytical workflow is presented in Figure 1. To harmonize postoperative timepoints across datasets with different sampling schemes, we first evaluated whether the modified ultrafiltration (MUF) [32,33] procedure in GSE143780 introduced a distinct transcriptional state. A random-gene heatmap and sample-sample correlation analysis showed that CPBend and MUF0h exhibited highly similar expression profiles and did not form separate sample clusters (Supplementary Figure 1a-d), indicating that MUF0h does not represent a transcriptionally distinct state relative to CPBend. Based on this observation, MUF8h and MUF24h were treated as the CPB8h and CPB24h timepoint in subsequent analyses.\u003c/p\u003e\n\u003cp\u003eAfter aligning all datasets to a unified axis consisting of PreCPB, CPBend (or Post), CPB8h, and CPB24h, we compared early transcriptional responses to CPB across the three transcriptomic datasets. We first summarized the sampling timepoints of whole-blood bulk RNA-seq (GSE143780), PBMC single-cell RNA-seq (GSE262146), and neutrophil bulk RNA-seq (GSE297377) (Figure 2a). Time-course clustering of GSE143780 revealed several distinct temporal expression patterns, including gene groups that were transiently induced, progressively increased, or gradually downregulated after CPB (Figure 2b). Consistently, all three datasets demonstrated widespread differential gene expression when comparing PreCPB and CPBend samples: large sets of upregulated and downregulated genes were detected in GSE143780 (CPB8h vs Pre), GSE262146 (CPB8h vs Pre), and GSE297377 (Post vs Pre), as shown in the heatmaps (Figure 2c-e) and volcano plots (Figure 2f-h). The full differential expression profiles at CPBend and CPB24h in GSE143780, and at Post and CPB24h in GSE262146, are presented for reference (Supplementary Figure 1e-l).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTemporal Pathway Dynamics in Whole Blood After CPB\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo characterize global functional responses to CPB, we performed GOBP and KEGG enrichment analyses across the three postoperative timepoints in GSE143780 (CPBend, CPB8h, and CPB24h). Upregulated pathways were dominated by innate immune and myeloid-related programs, including positive regulation of innate immune response, myeloid leukocyte activation, positive regulation of cytokine production, regulation of inflammatory response and leukocyte chemotaxis. KEGG enrichment further highlighted activation of Toll-like receptor signaling, platelet activation, hematopoietic cell lineage and focal adhesion. In contrast, downregulated GOBP terms were predominantly linked to lymphocyte effector functions, including cell killing, leukocyte-mediated cytotoxicity, leukocyte-mediated immunity and decreased positive regulation of NK cell-mediated immunity and cytotoxicity. KEGG analyses revealed suppression of antigen processing and presentation, along with additional immune-related modules (Figure 3a-d). GSEA was performed for the CPB8h timepoint. Upregulated signatures included monocyte activation, myeloid leukocyte activation, neutrophil degranulation and regulation of chemokine production whereas downregulated signatures involved CD4\u003csup\u003e+\u003c/sup\u003e T-cell proliferation, positive regulation of cell killing and NK-cell-mediated cytotoxicity (Figure 3e-f).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eLineage-specific Pathway Dynamics in PBMCs after CPB\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo determine the cellular sources underlying the whole-blood signatures, we examined lineage-specific transcriptional responses using the PBMC single-cell dataset (GSE262146). Major immune lineages including T cells, B cells, NK cells, monocyte-macrophage lineage cells, dendritic cells, progenitors, and cycling cells were annotated using canonical markers (Supplementary Figure 2a). UMAP visualization and compositional changes revealed a marked early expansion of monocyte-macrophage populations after CPB, whereas lymphocyte subsets exhibited a transient postoperative decrease followed by partial recovery at 24 h (Figure 4a-b). Differential expression analyses summarized the sets of upregulated and downregulated genes for each lineage across postoperative timepoints (Figure 4c-d), providing an overview of transcriptional perturbations in individual immune subsets. Monocyte-macrophage lineage cells displayed enrichment of innate and inflammatory programs. Upregulated GOBP terms included myeloid leukocyte activation, regulation of inflammatory response, and response to wounding, while KEGG analysis prominently highlighted complement and coagulation cascades along with additional inflammation-related pathways (Figure 4e; Supplementary Figure 2b). Downregulated pathways in this lineage mapped to broadly shared adaptive immune effector modules, including antigen processing and presentation and T-cell receptor signaling. T cells exhibited modest early upregulation of inflammatory and chemotactic responses, but key adaptive pathways including hematopoietic cell lineage, IL-17 signaling and T-cell activation were downregulated (Supplementary Figure 2c-d). B cells showed consistent reduction of antigen presentation, immune effector processes and broader adaptive immune regulatory modules that are annotated under T-cell receptor signaling and Th1/Th2/Th17 differentiation (Supplementary Figure 2e-k). NK cells demonstrated suppression of cytotoxicity-related pathways and immune-activating receptor signaling (Supplementary Figure 2g). As neutrophils are not captured within the PBMC compartment, we assessed their lineage-specific transcriptional responses using the neutrophil bulk RNA-seq dataset (GSE297377), which contains sorted neutrophil samples and therefore provides a reasonable approximation of lineage-specific signals not captured in PBMCs. Neutrophils exhibited activation of innate immune pathways, including IL-1\u0026beta; production, myeloid leukocyte activation, regulation of innate immune response, and MAPK signaling (Figure 4f-g). Only a small number of pathways, mainly related to apoptosis and leukocyte differentiation, were modestly downregulated.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eWGCNA Identifies a CPB-responsive Gene Module\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo identify coordinated gene programs underlying early transcriptional responses to CPB, we performed weighted gene co-expression network analysis (WGCNA) using the whole-blood bulk RNA-seq dataset (GSE143780). Sample clustering revealed no outlier samples across the four timepoints (PreCPB, CPBend, CPB8h, CPB24h), supporting the suitability of the dataset for network construction (Figure 5a). Soft-thresholding analysis identified an appropriate power that achieved scale-free topology while preserving network connectivity (Figure 5b). Hierarchical clustering of module eigengenes and module-module correlation patterns showed distinct co-expression modules with varying degrees of inter-module similarity (Figure 5c-d). Module-trait correlation analysis identified the\u003cem\u003e\u0026nbsp;lightcyan\u003c/em\u003e module as the most strongly associated with the CPB8h timepoint (correlation coefficient \u0026gt; 0.8), with weaker or opposite correlations at PreCPB, CPBend, and CPB24h (Figure 5e). Topological overlap matrix (TOM) visualization showed high intramodular connectivity within the \u003cem\u003elightcyan\u003c/em\u003e module (Figure 5f). Strong positive correlation between module membership and gene significance was observed, with highly connected genes also showing strong association with CPB8h (Figure 5g). Patient-level trajectories of the \u003cem\u003elightcyan\u003c/em\u003e module eigengene showed a consistent temporal pattern across individuals, with minimal change at CPBend, a pronounced peak at 8h, and partial decline by 24h (Figure 5h). \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eIntegrative Analysis and Machine Learning Select a Three-gene CPB Signature\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe integrated gene sets derived from single-cell, whole-blood bulk, and neutrophil bulk transcriptomic analyses. GeneSet1 consisted of genes shared between whole-blood bulk RNA-seq (GSE143780) and PBMC single-cell data (GSE262146), while GeneSet2 captured genes overlapping between whole blood and neutrophil bulk RNA-seq (GSE297377). GeneSet3 represented all genes within the CPB-associated \u003cem\u003elightcyan\u0026nbsp;\u003c/em\u003eWGCNA module. GeneSet4 was defined as the intersection between the combined GeneSet1 and GeneSet2 and GeneSet3, representing genes supported by both cross-dataset overlap and co-expression network analysis (Figure 6a). A protein-protein interaction (PPI) network was constructed using the combined GeneSet1 and GeneSet2, and the top 30 hub genes were identified using STRING and Cytoscape (Figure 6b). These hub genes, together with the WGCNA-derived GeneSet4, were subsequently evaluated using three machine-learning models including LASSO, Elastic Net, and Random Forest (Figure 6c). The intersection of features selected by all three models yielded three genes: cluster of differentiation 163 (\u003cem\u003eCD163\u003c/em\u003e), interleukin-10 (\u003cem\u003eIL10\u003c/em\u003e) and peroxisome proliferator-activated receptor gamma (\u003cem\u003ePPARG\u003c/em\u003e) (Figure 6d). Density plots showed that this three-gene model effectively separated CPB8h samples from PreCPB/ CPBend samples (Figure 6e). In addition, temporal profiling revealed that all three genes were markedly induced at CPB8h and partially normalized by CPB24h (Figure 6f).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eClinical Validation of The Three-gene CPB Signature and Its Association with Postoperative\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eComplications\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe next measured the expression levels of \u003cem\u003eCD163\u003c/em\u003e, \u003cem\u003eIL10\u003c/em\u003e and \u003cem\u003ePPARG\u003c/em\u003e in whole-blood samples at the timepoint of CPB24h. A total of 42 individuals who underwent cardiac surgery necessitating CPB and fulfilled the inclusion criteria were included, and their clinical characteristics and cardiac-lesion types were shown in Table 1 and Table 2, respectively. We assessed whether the signature was associated with postoperative complications, focusing on acute lung injury (ALI). Expression levels of all three genes were significantly higher in patients who developed ALI than in those without ALI (Figure 7a-c). Correlation analyses were performed between gene expression levels and the oxygenation index (OI). All three genes displayed significant correlations with OI (Figure 7d-f, \u003cem\u003eCD163\u003c/em\u003e: r = 0.6523, \u003cem\u003eP\u003c/em\u003e ＜ 0.0001; \u003cem\u003eIL10\u003c/em\u003e: r = 0.6397, \u003cem\u003eP\u0026nbsp;\u003c/em\u003e＜ 0.0001; \u003cem\u003ePPARG\u003c/em\u003e: r = 0.4935, \u003cem\u003eP\u003c/em\u003e = 0.0009). Meanwhile, we found that elevated \u003cem\u003eCD163\u0026nbsp;\u003c/em\u003e(r = 0.5068, \u003cem\u003eP\u0026nbsp;\u003c/em\u003e= 0.0006) and \u003cem\u003eIL10\u0026nbsp;\u003c/em\u003e(r = 0.5659, \u003cem\u003eP\u0026nbsp;\u003c/em\u003e＜ 0.0001) level were significantly correlated with increased hospital length of stay (LOS) (Figure 7g-h). PPARG levels also exhibited a positive, though non-significant, correlation with LOS (Figure 7i, r = 0.3010, \u003cem\u003eP\u003c/em\u003e = 0.0527).\u0026nbsp;\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this study, we integrated multi-layer transcriptomic datasets [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e] with network and machine-learning analyses to delineate early immune responses to CPB and identify a reproducible three-gene signature (\u003cem\u003eCD163\u003c/em\u003e, \u003cem\u003eIL10\u003c/em\u003e, \u003cem\u003ePPARG\u003c/em\u003e) that captures acute postoperative immune activation. Clinical validation further demonstrated that this signature is robustly induced in pediatric patients undergoing CPB and is associated with postoperative acute lung injury, underscoring its translational relevance.\u003c/p\u003e \u003cp\u003eAt the systems level, our analyses revealed a coherent pattern in which innate immune and myeloid programs dominate the early postoperative phase, whereas adaptive immunity is consistently attenuated. These findings align with the recognized role of CPB in triggering systemic inflammatory responses and leukocyte activation but extend previous observations by providing cell-type-resolved transcriptomic evidence across multiple datasets. The convergence of whole-blood enrichment patterns with monocyte-macrophage and neutrophil signatures in both single-cell and neutrophil data strongly suggests that myeloid cells constitute the primary drivers of acute inflammatory activation following CPB. These system-level patterns provided the biological rationale for identifying compact and reproducible molecular signatures that capture the core CPB-induced immune response. From a scientific perspective, our study addresses a major limitation of prior transcriptomic analyses, which often examined individual datasets or isolated leukocyte subsets. By integrating orthogonal data modalities, we identify reproducible gene programs that are robust across datasets. WGCNA enables the extraction of a patient-consistent co-expression module that peaked at the CPB8h timepoint, while network and machine-learning analyses further ensure the selected genes are not only differentially expressed but also represent central regulators of the CPB response. Together, these multi-layer analyses converge on a coherent three-gene signature that remains reproducible across datasets, analytical methods, and modeling approaches.\u003c/p\u003e \u003cp\u003eOn the basis of these integrative and reproducible transcriptomic patterns, we next sought to interpret the biological mechanisms underlying the identified three-gene signature. Previous studies indicated that the systemic inflammatory response syndrome induced by CPB is commonly accompanied by a compensatory anti-inflammatory response. This shift can result in an immunosuppressive state known as immunoparalysis, which leaves patients vulnerable to infections, a leading cause of morbidity after cardiac surgery [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e] [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. Neutrophils are typically among the earliest responder cells following CPB and activated neutrophils release proteolytic enzymes, reactive oxygen species, and multiple pro-inflammatory mediators, thereby exacerbating systemic inflammation and contributing to tissue injury [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e] [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. Neutrophil-driven innate immune activation after CPB was reported to be related to postoperative organ dysfunction including pulmonary complications[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e], and is increasingly considered as a hallmark of CPB-associated immune dysregulation. At the molecular level, neutrophils display robust activation of innate immune and inflammatory signaling pathways, characterized by upregulation of cytokine production, MAPK signaling, and degranulation-related programs [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e], highlighting their central role in initiating and amplifying postoperative inflammatory responses.\u003c/p\u003e \u003cp\u003eNeutrophil-driven inflammation, endothelial activation, and tissue injury signals subsequently provide critical upstream cues for monocyte-macrophage immune responses. In this context, the monocyte-macrophage lineage undergoes a transcriptional reprogramming dominated by immunoregulatory features. CD163, a macrophage-specific hemoglobin scavenger receptor, is markedly upregulated in response to CPB-associated hemolysis and systemic inflammatory stimuli, reflecting macrophage sensing of hemoglobin-heme stress and a shift toward an immunomodulatory phenotype [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]. Concurrently, induction of IL-10 represents a negative feedback mechanism activated to restrain excessive inflammation; however, sustained elevation of IL-10 may also contribute to postoperative immunosuppression [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]. In addition, activation of PPARG-associated transcriptional programs suggests metabolic-immune coupling within macrophages, promoting alternative macrophage polarization and suppression of pro-inflammatory signaling pathways [\u003cspan additionalcitationids=\"CR40\" citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eTaken together, CPB-induced immune responses can be conceptualized as a sequential process characterized by neutrophil-driven inflammatory initiation followed by macrophage-mediated immunoregulatory adaptation. In a subset of patients, this compensatory immunoregulatory program may become exaggerated, resulting in impaired innate immune defense and dysregulated tissue repair. In the present study, the expression levels of \u003cem\u003eCD163\u003c/em\u003e, \u003cem\u003eIL10\u003c/em\u003e, and \u003cem\u003ePPARG\u003c/em\u003e were significantly correlated with worsened oxygenation indices and extended hospital length of stay, suggesting that this immune phenotype may contribute to the development of postoperative acute lung injury.\u003c/p\u003e \u003cp\u003eThis study has several limitations. First, the public datasets used for discovery had relatively small sample sizes, a common limitation in all transcriptomic studies. Nonetheless, the convergence of findings across independent datasets and analytical frameworks supports the robustness of the three-gene signature. Second, although we validated gene expression levels in postoperative clinical samples, we did not assess their paired expression changes before and after CPB, which limits our ability to directly confirm perioperative transcriptional dynamics in our patients. Third, the three-gene signature was identified using transcriptomic profiles at the CPB8h timepoint in public datasets, whereas clinical validation was performed with CPB24h samples due to the availability of clinical specimens. This temporal mismatch may limit direct comparability of transcriptional dynamics across cohorts. Fourth, the clinical validation cohort was modest in size and derived from a single center. The predictive potential of the signature requires evaluation in larger and ideally prospective multicenter cohorts. Lastly, our analysis focused on early postoperative phases; longer-term immune trajectories remain to be explored. Further validation is essential to define the clinical utility of this signature.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn summary, this multi-layer transcriptomic analysis defines the immune architecture of early CPB responses and identifies a robust three-gene signature with both biological and clinical relevance. These findings highlight the central role of myeloid activation in pediatric CPB and suggest that the three-gene signature may serve as a useful biomarker for early risk stratification in pediatric CPB although subjected to further validation.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAuthor Contributions:\u0026nbsp;\u003c/strong\u003eZ.L. and X.W. contributed equally to this work. Z.L. contributed to data curation, formal analysis, visualization, methodology, interpretation of results, and writing of the original draft. X.W. contributed to data curation, validation, visualization, methodology, interpretation of results, and writing of the original draft. X.Z., J.F., L.Y., and H.Y. contributed to validation, data acquisition, and resources. D.M. contributed to methodological guidance and writing\u0026mdash;review and editing. L.Yu. and X.L. contributed to writing\u0026mdash;review and editing. Q.S. contributed to conceptualization, supervision, project administration, and writing\u0026mdash;review and editing. All authors have read and approved the final manuscript and agree to be accountable for their respective contributions.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding:\u0026nbsp;\u003c/strong\u003eThis work was supported by the Central Guiding Fund for Local Science and Technology Development Projects (No. 2023ZY1058).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eInstitutional Review Board Statement:\u0026nbsp;\u003c/strong\u003eThe study protocol was approved by the Ethics Committee of Children\u0026rsquo;s Hospital of Zhejiang University (Approval No. 2022-IRB-0054). All procedures were conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from the legal guardians of all participants for clinical patients\u0026rsquo; sample analysis part.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Availability Statement:\u0026nbsp;\u003c/strong\u003eAll transcriptomic datasets analyzed in this study were obtained from the Gene Expression Omnibus (GEO) under accession numbers GSE143780 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE143780), GSE262146 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE262146), and GSE297377 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE297377). No new raw sequencing data were generated in this study, and all analyses were performed using publicly available datasets with standard R packages. No additional custom code or proprietary software was created beyond the scripts described in the Methods section.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments:\u003c/strong\u003e The authors thank all patients and their families for their participation in this study. We also thank the clinical staff of the Department of Pediatric Cardiac Surgery, Children\u0026rsquo;s Hospital, Zhejiang University School of Medicine, for their assistance with sample collection and clinical data acquisition.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflicts of Interests:\u0026nbsp;\u003c/strong\u003eThe authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eBacker, C.L.; Overman, D.M.; Dearani, J.A.; Romano, J.C.; Tweddell, J.S.; Kumar, S.R.; Marino, B.S.; Bacha, E.A.; Jaquiss, R.D.B.; Zaidi, A.N.; et al. Recommendations for centers performing pediatric heart surgery in the United States. \u003cem\u003eThe Journal of thoracic and cardiovascular surgery \u003c/em\u003e\u003cstrong\u003e2023\u003c/strong\u003e, \u003cem\u003e166\u003c/em\u003e, 1782-1820, doi:10.1016/j.jtcvs.2023.09.001.\u003c/li\u003e\n\u003cli\u003eKumar, S.R. et al. \u003cem\u003eThe Society of Thoracic Surgeons Congenital Heart Surgery Database: 2022 Update on Outcomes and Research. Ann Thorac Surg 115, 807-819 (2023). \u003c/em\u003e\u003cstrong\u003e2023\u003c/strong\u003e.\u003c/li\u003e\n\u003cli\u003eLiu, H.; Hu, Y.J.; Zheng, S.Q.; Chen, T.; Zeng, Z.H.; Wu, D.D.; Zhao, S.; Zeng, B.; Liu, Z.G.; Shao, Y.F. Effect of perfusate oxygenation on inflammatory response in congenital heart disease children from low versus high altitude. \u003cem\u003eThe Journal of thoracic and cardiovascular surgery \u003c/em\u003e\u003cstrong\u003e2021\u003c/strong\u003e, \u003cem\u003e161\u003c/em\u003e, 2180-2190, doi:10.1016/j.jtcvs.2020.05.108.\u003c/li\u003e\n\u003cli\u003eEsper, S.A.; Subramaniam, K.; Tanaka, K.A. Pathophysiology of Cardiopulmonary Bypass: Current Strategies for the Prevention and Treatment of Anemia, Coagulopathy, and Organ Dysfunction. \u003cem\u003eSemin Cardiothorac Vasc Anesth \u003c/em\u003e\u003cstrong\u003e2014\u003c/strong\u003e, \u003cem\u003e18\u003c/em\u003e, 161-176, doi:10.1177/1089253214532375.\u003c/li\u003e\n\u003cli\u003eMahle, W.T.; Matthews, E.; Kanter, K.R.; Kogon, B.E.; Hamrick, S.E.; Strickland, M.J. Inflammatory response after neonatal cardiac surgery and its relationship to clinical outcomes. \u003cem\u003eAnn Thorac Surg \u003c/em\u003e\u003cstrong\u003e2014\u003c/strong\u003e, \u003cem\u003e97\u003c/em\u003e, 950-956, doi:10.1016/j.athoracsur.2013.10.069.\u003c/li\u003e\n\u003cli\u003eAppachi; E; Mossad; E; Mee; R.B. \u0026amp; Bokesch, P. \u003cem\u003ePerioperative serum interleukins in neonates with hypoplastic left-heart syndrome and transposition of the great arteries. J Cardiothorac Vasc Anesth 21, 184-190 (2007). \u003c/em\u003e\u003cstrong\u003e2007\u003c/strong\u003e.\u003c/li\u003e\n\u003cli\u003eLesouhaitier, M.; Belicard, F.; Tadi\u0026eacute;, J.M. Cardiopulmonary bypass and VA-ECMO induced immune dysfunction: common features and differences, a narrative review. \u003cem\u003eCritical care (London, England) \u003c/em\u003e\u003cstrong\u003e2024\u003c/strong\u003e, \u003cem\u003e28\u003c/em\u003e, 300, doi:10.1186/s13054-024-05058-z.\u003c/li\u003e\n\u003cli\u003eTu, L.N.; Hsieh, L.; Kajimoto, M.; Charette, K.; Kibiryeva, N.; Forero, A.; Hampson, S.; Marshall, J.A.; O\u0026apos;Brien, J.; Scatena, M.; et al. Shear stress associated with cardiopulmonary bypass induces expression of inflammatory cytokines and necroptosis in monocytes. \u003cem\u003eJCI insight \u003c/em\u003e\u003cstrong\u003e2021\u003c/strong\u003e, \u003cem\u003e6\u003c/em\u003e, doi:10.1172/jci.insight.141341.\u003c/li\u003e\n\u003cli\u003eJain, P.N.; Robertson, M.; Lasa, J.J.; Shekerdemian, L.; Guffey, D.; Zhang, Y.; Lingappan, K.; Checchia, P.; Coarfa, C. Altered metabolic and inflammatory transcriptomics after cardiac surgery in neonates with congenital heart disease. \u003cem\u003eSci Rep \u003c/em\u003e\u003cstrong\u003e2021\u003c/strong\u003e, \u003cem\u003e11\u003c/em\u003e, 4965, doi:10.1038/s41598-021-83882-x.\u003c/li\u003e\n\u003cli\u003eLi, W.; Tu, L.N.; Hsieh, L.; Smith, J.R.; Yeh, Y.-T.; Sinyagin, A.; Ghassemian, M.; Timms, A.; Charette, K.; Mauchley, D.; et al. Cardiopulmonary bypass activates classical monocytes via shear-mediated activation of Store-Operated Calcium Entry. \u003cem\u003ebioRxiv \u003c/em\u003e\u003cstrong\u003e2024\u003c/strong\u003e, 2022.2005.2004.490549, doi:10.1101/2022.05.04.490549.\u003c/li\u003e\n\u003cli\u003eChen, H.; Liu, J.; Wu, Y.; Jiang, L.; Tang, M.; Wang, X.; Fang, X.; Wang, X. Weighted gene co-expression identification of CDKN1A as a hub inflammation gene following cardiopulmonary bypass in children with congenital heart disease. \u003cem\u003eFront Surg \u003c/em\u003e\u003cstrong\u003e2022\u003c/strong\u003e, \u003cem\u003e9\u003c/em\u003e, 963850, doi:10.3389/fsurg.2022.963850.\u003c/li\u003e\n\u003cli\u003eMaisat W, H.L.S.S.S.Y.C.K.S.V.P.H. Neutrophil extracellular traps formation is associated with postoperative complications in congenital cardiac surgery. \u003cem\u003ePediatr Res. (2024):1\u0026ndash;14. doi:10.1038/s41390-024-03717-z \u003c/em\u003e\u003cstrong\u003e2024\u003c/strong\u003e.\u003c/li\u003e\n\u003cli\u003eMaisat, W.; Sandhu, S.; Kim, S.; Van Pelt, H.; Kong, S.W.; Ibla, J.; Yuki, K. Neutrophil Myo5c gene downregulation is associated with postoperative organ dysfunction following pediatric cardiac surgery with cardiopulmonary bypass. \u003cem\u003eFront Cardiovasc Med \u003c/em\u003e\u003cstrong\u003e2025\u003c/strong\u003e, \u003cem\u003e12\u003c/em\u003e, 1380606, doi:10.3389/fcvm.2025.1380606.\u003c/li\u003e\n\u003cli\u003eRaggi, F.; Cangelosi, D.; Becherini, P.; Blengio, F.; Morini, M.; Acquaviva, M.; Belli, M.L.; Panizzon, G.; Cervo, G.; Varesio, L.; et al. Transcriptome analysis defines myocardium gene signatures in children with ToF and ASD and reveals disease-specific molecular reprogramming in response to surgery with cardiopulmonary bypass. \u003cem\u003eJ Transl Med \u003c/em\u003e\u003cstrong\u003e2020\u003c/strong\u003e, \u003cem\u003e18\u003c/em\u003e, 21, doi:10.1186/s12967-020-02210-5.\u003c/li\u003e\n\u003cli\u003eWang, Y.; Chen, L.; Yao, C.; Wang, T.; Wu, J.; Shang, Y.; Li, B.; Xia, H.; Huang, S.; Wang, F.; et al. Early plasma proteomic biomarkers and prediction model of acute respiratory distress syndrome after cardiopulmonary bypass: a prospective nested cohort study. \u003cem\u003eInternational journal of surgery (London, England) \u003c/em\u003e\u003cstrong\u003e2023\u003c/strong\u003e, \u003cem\u003e109\u003c/em\u003e, 2561-2573, doi:10.1097/js9.0000000000000434.\u003c/li\u003e\n\u003cli\u003eLove, M.I.; Huber, W.; Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. \u003cem\u003eGenome Biology \u003c/em\u003e\u003cstrong\u003e2014\u003c/strong\u003e, \u003cem\u003e15\u003c/em\u003e, 550-550.\u003c/li\u003e\n\u003cli\u003eWickham, H. ggplot2: Elegant Graphics for Data Analysis. \u003cem\u003eSpringer \u003c/em\u003e\u003cstrong\u003e2016\u003c/strong\u003e.\u003c/li\u003e\n\u003cli\u003eGu, Z.; Eils, R.; Schlesner, M. Complex heatmaps reveal patterns and correlations in multidimensional genomic data. \u003cem\u003eBioinformatics \u003c/em\u003e\u003cstrong\u003e2016\u003c/strong\u003e, \u003cem\u003e32\u003c/em\u003e, 2847-2849.\u003c/li\u003e\n\u003cli\u003eStuart, T.; Butler, A.; Hoffman, P.; et al. Comprehensive Integration of Single-Cell Data. \u003cem\u003eCell \u003c/em\u003e\u003cstrong\u003e2019\u003c/strong\u003e, \u003cem\u003e177\u003c/em\u003e, 1888-1902.\u003c/li\u003e\n\u003cli\u003eAmezquita, R.A.; Lun, A.T.L.; Becht, E.; Carey, V.J.; Carpp, L.N.; Geistlinger, L.; Marini, F.; Rue-Albrecht, K.; Risso, D.; Soneson, C.; et al. Orchestrating single-cell analysis with Bioconductor. \u003cem\u003eNat Methods \u003c/em\u003e\u003cstrong\u003e2020\u003c/strong\u003e, \u003cem\u003e17\u003c/em\u003e, 137-145, doi:10.1038/s41592-019-0654-x.\u003c/li\u003e\n\u003cli\u003eXu, S.; Hu, E.; Cai, Y.; Xie, Z.; Luo, X.; Zhan, L.; Tang, W.; Wang, Q.; Liu, B.; Wang, R.; et al. Using clusterProfiler to characterize multiomics data. \u003cem\u003eNat Protoc \u003c/em\u003e\u003cstrong\u003e2024\u003c/strong\u003e, \u003cem\u003e19\u003c/em\u003e, 3292-3320, doi:10.1038/s41596-024-01020-z.\u003c/li\u003e\n\u003cli\u003eLiberzon, A.; Birger, C.; Thorvaldsdottir, H.; Ghandi, M.; Mesirov, J.P.; Tamayo, P. The Molecular Signatures Database (MSigDB) hallmark gene set collection. \u003cem\u003eCell Syst \u003c/em\u003e\u003cstrong\u003e2015\u003c/strong\u003e, \u003cem\u003e1\u003c/em\u003e, 417-425, doi:10.1016/j.cels.2015.12.004.\u003c/li\u003e\n\u003cli\u003eYu, G.; He, Q.Y. ReactomePA: an R/Bioconductor package for reactome pathway analysis and visualization. \u003cem\u003eMolecular BioSystems \u003c/em\u003e\u003cstrong\u003e2016\u003c/strong\u003e, \u003cem\u003e12\u003c/em\u003e, 477-479.\u003c/li\u003e\n\u003cli\u003eKanehisa, M.; Furumichi, M.; Sato, Y.; Kawashima, M.; Ishiguro-Watanabe, M. KEGG for taxonomy-based analysis of pathways and genomes. \u003cem\u003eNucleic Acids Res \u003c/em\u003e\u003cstrong\u003e2023\u003c/strong\u003e, \u003cem\u003e51\u003c/em\u003e, D587-d592, doi:10.1093/nar/gkac963.\u003c/li\u003e\n\u003cli\u003eLangfelder, P.; Horvath, S. WGCNA: an R package for weighted correlation network analysis. \u003cem\u003eBMC Bioinformatics \u003c/em\u003e\u003cstrong\u003e2008\u003c/strong\u003e, \u003cem\u003e9\u003c/em\u003e, 559-559.\u003c/li\u003e\n\u003cli\u003eSzklarczyk, D.; et al. The STRING database in 2021: customizable protein\u0026ndash;protein networks and functional characterization. \u003cem\u003eNucleic Acids Research \u003c/em\u003e\u003cstrong\u003e2021\u003c/strong\u003e, \u003cem\u003e49\u003c/em\u003e, D605-D612.\u003c/li\u003e\n\u003cli\u003eShannon, P.; et al. Cytoscape: a software environment for integrated models of biomolecular interaction networks. \u003cem\u003eGenome Research \u003c/em\u003e\u003cstrong\u003e2003\u003c/strong\u003e, \u003cem\u003e13\u003c/em\u003e, 2498-2504.\u003c/li\u003e\n\u003cli\u003eChin, C.H.; et al. cytoHubba: identifying hub objects and sub-networks from complex interactome. \u003cem\u003eBMC Systems Biology \u003c/em\u003e\u003cstrong\u003e2014\u003c/strong\u003e, \u003cem\u003e8\u003c/em\u003e, S11-S11.\u003c/li\u003e\n\u003cli\u003eFriedman, J.; Hastie, T.; Tibshirani, R. Regularization Paths for Generalized Linear Models via Coordinate Descent. \u003cem\u003eJournal of Statistical Software \u003c/em\u003e\u003cstrong\u003e2010\u003c/strong\u003e, \u003cem\u003e33\u003c/em\u003e, 1-22.\u003c/li\u003e\n\u003cli\u003eBreiman, L. Random forests. \u003cem\u003eMachine Learning \u003c/em\u003e\u003cstrong\u003e2001\u003c/strong\u003e, \u003cem\u003e45\u003c/em\u003e, 5-32.\u003c/li\u003e\n\u003cli\u003eEmeriaud, G.; L\u0026oacute;pez-Fern\u0026aacute;ndez, Y.M.; Iyer, N.P.; Bembea, M.M.; Agulnik, A.; Barbaro, R.P.; Baudin, F.; Bhalla, A.; Brunow de Carvalho, W.; Carroll, C.L.; et al. Executive Summary of the Second International Guidelines for the Diagnosis and Management of Pediatric Acute Respiratory Distress Syndrome (PALICC-2). \u003cem\u003ePediatr Crit Care Med \u003c/em\u003e\u003cstrong\u003e2023\u003c/strong\u003e, \u003cem\u003e24\u003c/em\u003e, 143-168, doi:10.1097/pcc.0000000000003147.\u003c/li\u003e\n\u003cli\u003eAd, N.; Snir, E.; Katz, J.; Birk, E.; Vidne, B.A. Use of the modified technique of ultrafiltration in pediatric open-heart surgery: a prospective study. \u003cem\u003eIsr J Med Sci \u003c/em\u003e\u003cstrong\u003e1996\u003c/strong\u003e, \u003cem\u003e32\u003c/em\u003e, 1326-1331.\u003c/li\u003e\n\u003cli\u003eChew, M.S.; Brix-Christensen, V.; Ravn, H.B.; Brandslund, I.; Ditlevsen, E.; Pedersen, J.; Hjortholm, K.; Hansen, O.K.; Tonnesen, E.; Hjortdal, V.E. Effect of modified ultrafiltration on the inflammatory response in paediatric open-heart surgery: a prospective, randomized study. \u003cem\u003ePerfusion \u003c/em\u003e\u003cstrong\u003e2002\u003c/strong\u003e, \u003cem\u003e17\u003c/em\u003e, 327-333, doi:10.1191/0267659102pf595oa.\u003c/li\u003e\n\u003cli\u003eLiu, C.H.; Huang, Z.H.; Huang, S.C.; Jou, T.S. Endocytosis of peroxiredoxin 1 links sterile inflammation to immunoparalysis in pediatric patients following cardiopulmonary bypass. \u003cem\u003eRedox biology \u003c/em\u003e\u003cstrong\u003e2021\u003c/strong\u003e, \u003cem\u003e46\u003c/em\u003e, 102086, doi:10.1016/j.redox.2021.102086.\u003c/li\u003e\n\u003cli\u003eChew, M.S.; Brandslund, I.; Brix-Christensen, V.; Ravn, H.B.; Hjortdal, V.E.; Pedersen, J.; Hjortdal, K.; Hansen, O.K.; Tonnesen, E. Tissue injury and the inflammatory response to pediatric cardiac surgery with cardiopulmonary bypass: a descriptive study. \u003cem\u003eAnesthesiology \u003c/em\u003e\u003cstrong\u003e2001\u003c/strong\u003e, \u003cem\u003e94\u003c/em\u003e, 745-753; discussion 745A, doi:10.1097/00000542-200105000-00010.\u003c/li\u003e\n\u003cli\u003eEtzerodt, A.; Moestrup, S.K. CD163: a scavenger receptor clearing hemoglobin\u0026ndash;haptoglobin complexes. \u003cem\u003eImmunology \u003c/em\u003e\u003cstrong\u003e2013\u003c/strong\u003e.\u003c/li\u003e\n\u003cli\u003ePhilippidis, P.; Mason, J.C.; Evans, B.J.; Nadra, I.; Taylor, K.M.; Haskard, D.O.; Landis, R.C. Hemoglobin scavenger receptor CD163 mediates interleukin-10 release and heme oxygenase-1 synthesis: antiinflammatory monocyte-macrophage responses in vitro, in resolving skin blisters in vivo, and after cardiopulmonary bypass surgery. \u003cem\u003eCirculation research \u003c/em\u003e\u003cstrong\u003e2004\u003c/strong\u003e, \u003cem\u003e94\u003c/em\u003e, 119-126, doi:10.1161/01.Res.0000109414.78907.F9.\u003c/li\u003e\n\u003cli\u003eMoore, K.W.; et al. IL-10 and the regulation of immune responses. \u003cem\u003eAnnual Review of Immunology \u003c/em\u003e\u003cstrong\u003e2001\u003c/strong\u003e.\u003c/li\u003e\n\u003cli\u003eOdegaard, J.I.; Ricardo-Gonzalez, R.R.; Goforth, M.H.; Morel, C.R.; Subramanian, V.; Mukundan, L.; Red Eagle, A.; Vats, D.; Brombacher, F.; Ferrante, A.W.; et al. Macrophage-specific PPARgamma controls alternative activation and improves insulin resistance. \u003cem\u003eNature \u003c/em\u003e\u003cstrong\u003e2007\u003c/strong\u003e, \u003cem\u003e447\u003c/em\u003e, 1116-1120, doi:10.1038/nature05894.\u003c/li\u003e\n\u003cli\u003eBouhlel, M.A.; Derudas, B.; Rigamonti, E.; Di\u0026egrave;vart, R.; Brozek, J.; Haulon, S.; Zawadzki, C.; Jude, B.; Torpier, G.; Marx, N.; et al. PPARgamma activation primes human monocytes into alternative M2 macrophages with anti-inflammatory properties. \u003cem\u003eCell Metab \u003c/em\u003e\u003cstrong\u003e2007\u003c/strong\u003e, \u003cem\u003e6\u003c/em\u003e, 137-143, doi:10.1016/j.cmet.2007.06.010.\u003c/li\u003e\n\u003cli\u003eLiu, B.; Liang, G.; Xu, G.; Liu, D.; Cai, Q.; Gao, Z. Intervention of rosiglitazone on myocardium Glut-4 mRNA expression during ischemia-reperfusion injury in cardio-pulmonary bypass in dogs. \u003cem\u003eMolecular and cellular biochemistry \u003c/em\u003e\u003cstrong\u003e2013\u003c/strong\u003e, \u003cem\u003e373\u003c/em\u003e, 279-284, doi:10.1007/s11010-012-1501-x.\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003e\u003cstrong\u003eTable 1\u003c/strong\u003e\u003cstrong\u003e．\u003c/strong\u003e\u003cstrong\u003eDemographic and clinical characteristics\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eStudy cohort\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(n=42)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eALI\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(n=18)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eNon-ALI\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(n=24)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e\u003cem\u003eP\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eSex (male; n [%])\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e27(64.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e11(61.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e16(66.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.754\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eAge (months)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e8.69\u0026plusmn;1.626\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e4.83\u0026plusmn;0.933\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e11.58\u0026plusmn;2.63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.038\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eWeight (kg)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e7.2\u0026plusmn;0.451\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e6.483\u0026plusmn;0.448\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e7.375\u0026plusmn;0.709\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.193\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eOperation time (minutes)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e155.5\u0026plusmn;6.142\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e171.1\u0026plusmn;11.32\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e143.7\u0026plusmn;5.716\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.007\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eCPB time(minutes)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e88.93\u0026plusmn;5.359\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e98.72\u0026plusmn;10.68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e81.58\u0026plusmn;4.57\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.117\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eAC time (minutes)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e55.24\u0026plusmn;4.543\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e63.67\u0026plusmn;9.147\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e48.92\u0026plusmn;3.736\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.109\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eHospital LOS (days)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e16.79\u0026plusmn;2.180\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e18.61\u0026plusmn;1.684\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e15.42\u0026plusmn;3.614\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.0021\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eData are presented as number of patients (%), mean \u0026plusmn; SEM, or counts, as appropriate. ALI, acute lung injury; CPB, cardiopulmonary bypass; LOS, length of stay; AC, aortic cross-clamp.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2\u003c/strong\u003e\u003cstrong\u003e．\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;Cardiac lesion types\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 107px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 101px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eStudy cohort\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 94px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eALI\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 132px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eNon-ALI\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 107px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eVSD\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 101px;\"\u003e\n \u003cp\u003e20\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 94px;\"\u003e\n \u003cp\u003e10(50%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 132px;\"\u003e\n \u003cp\u003e10(50%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 107px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eASD\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 101px;\"\u003e\n \u003cp\u003e7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 94px;\"\u003e\n \u003cp\u003e1(14%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 132px;\"\u003e\n \u003cp\u003e6(86)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 107px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTOF\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 101px;\"\u003e\n \u003cp\u003e9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 94px;\"\u003e\n \u003cp\u003e5(56%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 132px;\"\u003e\n \u003cp\u003e4(44%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 107px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eASD plus VSD\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 101px;\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 94px;\"\u003e\n \u003cp\u003e1(20%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 132px;\"\u003e\n \u003cp\u003e4(80%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 107px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePS\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 101px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 94px;\"\u003e\n \u003cp\u003e1(100%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 132px;\"\u003e\n \u003cp\u003e0(1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eData are presented as counts (%). ASD, atrial septal defect; PS, pulmonary stenosis; TOF, tetralogy of Fallot; VSD, ventricular septal defect.\u003c/p\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":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Transcriptomics, Machine learning, Cardiopulmonary bypass, Pediatrics, Congenital heart disease, Acute lung injury","lastPublishedDoi":"10.21203/rs.3.rs-8751380/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8751380/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003ePediatric cardiopulmonary bypass (CPB) induces profound systemic inflammation and immune dysregulation that contribute to postoperative complications. However, conserved transcriptional programs underlying CPB-associated immune responses remain incompletely characterized. This study aimed to define reproducible immune transcriptional signatures associated with postoperative acute lung injury in pediatric CPB. Three complementary transcriptomic datasets, including whole-blood bulk RNA sequencing, peripheral blood mononuclear cell single-cell RNA sequencing, and neutrophil bulk RNA sequencing, were integrated to characterize immune responses. Differential expression, functional enrichment, co-expression network analysis, protein-protein interaction analysis, and machine-learning-based feature selection were applied to identify robust genes. Key candidates were validated in pediatric cohorts using quantitative real-time PCR. Cross-dataset integration revealed consistent activation of innate immune and myeloid programs after CPB. Integrated network and machine-learning analyses converged on a three-gene signature comprising \u003cem\u003eCD163\u003c/em\u003e, \u003cem\u003eIL10\u003c/em\u003e, and \u003cem\u003ePPARG\u003c/em\u003e. Clinical validation demonstrated significant postoperative upregulation of all three genes, which correlated with the oxygenation index, indicating an association with postoperative ALI severity. This integrative transcriptomic analysis identifies a reproducible three-gene immune signature associated with postoperative acute lung injury following pediatric CPB. These findings provide insight into CPB-induced immune dysregulation and support the potential relevance of compact immune-related gene signatures for early postoperative risk stratification.\u003c/p\u003e","manuscriptTitle":"Identification of biomarkers associated with acute lung injury after cardiopulmonary bypass by integrative transcriptomic analysis and clinical validation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-11 16:44:33","doi":"10.21203/rs.3.rs-8751380/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-05-05T06:29:33+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-04T09:21:36+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"71151132115144476671055775885629897332","date":"2026-04-29T06:34:12+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"250332637309214861564465230754089396149","date":"2026-04-29T02:16:23+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"71882816727030842806719415399963979961","date":"2026-04-28T23:53:36+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-26T15:00:35+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-21T11:20:31+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"273826366390329465667080067365874584887","date":"2026-02-14T15:43:03+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"317267757840643987837135175595237189942","date":"2026-02-09T20:39:12+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-02-06T13:01:43+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-02-06T12:56:16+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-02-06T10:20:54+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-02-06T06:31:24+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-02-06T06:15:10+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"a2396c0c-66f9-46f9-bc61-0a6aa63fe7b7","owner":[],"postedDate":"February 11th, 2026","published":true,"recentEditorialEvents":[{"type":"decision","content":"Revision requested","date":"2026-05-05T06:29:33+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-04T09:21:36+00:00","index":84,"fulltext":""},{"type":"reviewerAgreed","content":"71151132115144476671055775885629897332","date":"2026-04-29T06:34:12+00:00","index":82,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"in-revision","subjectAreas":[{"id":62618493,"name":"Health sciences/Biomarkers"},{"id":62618494,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":62618495,"name":"Health sciences/Diseases"},{"id":62618496,"name":"Biological sciences/Genetics"},{"id":62618497,"name":"Biological sciences/Immunology"},{"id":62618498,"name":"Health sciences/Medical research"}],"tags":[],"updatedAt":"2026-05-05T06:39:17+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-11 16:44:33","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8751380","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8751380","identity":"rs-8751380","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}