Multi‑omics reveal neutrophil heterogeneity in sepsis (Review)

Introduction

Sepsis is defined as life-threatening organ dysfunction resulting from a dysregulated host response to infection (1). Each year, an estimated 48.9 million people worldwide develop sepsis, leading to approximately 11 million deaths and accounting for nearly one-fifth of all global mortality (2,3). Sepsis is a challenging health problem worldwide with lingering sequelae (4-6). Sepsis is considered a pathway to death initiated by the host immune defense system failing to restore homeostasis in response to an invading pathogen (7-12). Understanding the potential molecular and cellular features involved in complex immune pathological mechanisms is key for sepsis management. Neutrophils, as the most abundant type of white blood cell, are essential frontline responders to infection (13). In different stages of sepsis, neutrophils exhibit different transcriptomic profiles and biological functions (14), providing phenotypical diversity. Neutrophil subpopulations have been classified with different subsets causing various pathophysiological changes in the complex environment of sepsis, such as immune dysregulation, coagulation dysfunction and organ damage (15).

Neutrophil heterogeneity in sepsis has been extensively investigated (16-21). In disagreement with the previous consensus of neutrophil homogeneity, studies have indicated that neutrophils may remain in the circulation long enough to interpret environmental signals and execute specific molecular programs, providing a rationale for neutrophil diversity in vivo and promoting neutrophil heterogeneity research (16-21). Researchers have identified different subsets of neutrophils via surface proteins, such as CD molecules (22-25). However, there are no uniform standards for different neutrophil subpopulations distinguished by phenotypes (26), resulting in confusing neutrophil nomenclature and disagreements regarding neutrophil function. It remains unclear whether neutrophil subsets present in sepsis result from bone marrow mobilization or are specific subsets formed under the influence of sepsis. Moreover, whether the neutrophil response to sepsis represents a transient and reversible activation, or profound gene reprogramming with long-lasting consequences for host immunity remains unknown. Ng et al (21) proposed that the integration of protein and transcript composition, functional properties, tissue distribution, and genomic organization provides a more definitive framework for classifying immune cells, thereby offering a more precise approach to defining true neutrophil heterogeneity (21). To define these features, multi-omics analyses provide a more comprehensive understanding of cellular heterogeneity.

Omics analyses evaluate the complete set of molecular data, comprising genome, transcriptome, proteome and metabolome (27). Single omics studies, predominantly single-cell RNA sequencing (scRNA-seq), have revealed notable heterogeneity of neutrophils (28-30). High-throughput analysis based on scRNA-seq enables broad screening of neutrophil heterogeneity, allowing precise identification of neutrophil subsets combined with more practical methods, such as flow cytometry (28). As a combination of single omics approaches, multi-omics unveils the information of a single regulatory layer and the association between different layers (31,32). Sepsis is heterogeneous and has a complicated immunological mechanism, resulting in modulation of neutrophil diversity (7,33). Multi-omics analyses can help to elucidate underlying differences of neutrophils in sepsis that have not been identified by previous classification methods.

The present review summarizes the phenotypical and functional heterogeneity of neutrophils, as well as advances in the discovery of neutrophil subsets in sepsis using single omics approaches and the complex biological mechanisms of neutrophils in sepsis using multi-omics analysis. The holistic perspective of multi-omics clarifies the dynamic interplay between neutrophils and the host immune response, and provides a foundation for the identification of novel biomarkers and therapeutic targets.

Neutrophil heterogeneity in sepsis

In sepsis, exposure to local or systemic extrinsic factors may modify neutrophil properties, resulting in the diversity of neutrophils (21). Neutrophil phenotypes and functions reflect key features of neutrophil heterogeneity (Fig. 1).

Figure 1 Phenotypical heterogeneity of neutrophils in homeostasis and sepsis. Under homeostatic conditions, immature neutrophils are characterized by expression of CD64, CD49d and CXCR4, whereas mature neutrophils typically express CD10, CD16b and CD35. During sepsis, both immature and mature neutrophils are exposed to inflammatory cues that reprogram cellular activities and functional states, leading to marked phenotypical shifts. Specifically, surface molecules such as CD11b, CD64, CD177, ICAM-1 and TREM-1 are upregulated, while CD10, CD62L, CD16 and CXCR2 are downregulated. These alterations reflect changes in neutrophil activation, adhesion and migration, however, the precise functional consequences and regulatory mechanisms underlying these transitions remain incompletely understood. CXCR, C-X-C chemokine receptor; ICAM-1, intercellular adhesion molecule-1; TREM-1, triggering receptor expressed on myeloid cells-1.

Phenotypical diversity of neutrophils in sepsis

Morphologically, the stages of neutrophil development include promyelocytes, myelocytes, metamyelocytes, band cells and segmented neutrophils, among which the segmented neutrophil is considered the mature form (34). Because of their characteristic multilobed nuclei at the mature stage, neutrophils are often referred to as polymorphonuclear leukocytes (PMNs). CD11b+ CD66b+ CD15+ CD14- are commonly used phenotypical markers for human neutrophils (22). In homeostasis, as neutrophils mature, neutrophil surface markers changed to facilitate altered function. Immature neutrophils express more C-X-C chemokine receptor type 4 (CXCR4) than mature neutrophils, which may promote its retention in bone marrow (23). CD16b, CD35 and CD10 appear after neutrophil maturation, whereas CD49d and CD64 disappear (24).

During sepsis, neutrophils present with phenotypical alterations. Seree-Aphinan et al (25) found that a decrease in CXCR2 surface levels is associated with sepsis due to internalization of the CXCR2 receptor induced by circulating chemokines (35,36). Other studies also support decreased expression of the chemokine receptor CXCR2 on the surface of neutrophils in patients with septic shock (37,38). In addition, CD11b is decreased, but CXCR1 expression does not change significantly (37,38). Demaret et al demonstrated that CD10dim CD16dim neutrophils have an increased frequency in patients with septic shock (39). Neutrophils exhibit increased CD11b expression and decreased CD62L expression following stimulation with IL-8 or N-formyl-methionyl-leucyl-pheny lalanine (39). Similarly, suppressor cells mobilized during acute inflammation are characterized by normal expression of CD16, low expression of CD62L and high expression of CD11b and CD11c (40). Geng et al identified a third immune regulatory neutrophil in different inflammation conditions (41). This hybrid neutrophil subset extravasates at sites of inflammation or infection and expresses dendritic cell markers CD11c, major histocompatibility complex class II and costimulatory molecules. Ode et al (42), using a mouse model of sepsis, found that the frequency and number of Intercellular adhesion molecule-1 positive neutrophils are increased. CD64, a high-affinity immunoglobulin Fcγ receptor I, mediates phagocytosis of bacteria. Under homeostatic conditions, the expression of CD64 on neutrophils is comparatively low (42). By contrast, during infection, proinflammatory cytokines induce a 10-fold increase in CD64 expression (43,44). In addition, CD64 expression on neutrophils is specific for bacterial infection (45,46). Triggering receptor expressed on myeloid cells-1, a member of the immunoglobulin superfamily, is upregulated when PMNs are exposed to bacteria (47). Demaret et al (48) revealed that the expression of CD177 mRNA and protein is increased in circulating neutrophils in sepsis.

When differentiating solely by markers, however, it remains unclear whether certain neutrophil subsets are true lineages or simply represent differentiation and maturation states induced by the tissue environment.

Functional heterogeneity of neutrophils in sepsis

As first-line immune cells, neutrophils are required to adapt to diverse environments and respond promptly. To address this challenge, neutrophils have multiple capabilities. Typical functions of neutrophils encompass granule generation and degranulation, secretion of antimicrobial proteins, production of reactive oxygen species (ROS), phagocytosis, and formation of neutrophil extracellular traps (NETs) (49). NETs are DNA scaffolds containing granule-derived proteins, such as enzymatically active proteases and anti-microbial peptides, formed to immobilize invading microorganisms or in response to sterile stimuli (50). Table I summarizes the functional diversity of neutrophils in sepsis (39,40,51-65).

Table I Heterogeneity of neutrophil functions in sepsis.

Table I

Heterogeneity of neutrophil functions in sepsis.

Functional aspect	Observations in sepsis	(Refs.)
NET formation (NETosis)	Decreased NETosis observed in septic patients; associated antimicrobial defense	(51)
Apoptosis regulation	Increased neutrophil production and decreased apoptosis in bone marrow; delayed apoptosis in circulating neutrophils	(39,51,64)
Phagocytosis	Enhanced phagocytic function in septic shock; immature neutrophils show decreased phagocytosis	(39,52,64-66)
Oxidative burst/ROS production	Upregulated ROS generation early in sepsis; decreased oxidative burst in late sepsis; excessive ROS may drive tissue injury	(39,40,56,58-61)
Chemotaxis/migration	Strongly inhibited chemotaxis; defective migration in septic patients; mediated by GRK activation via LPS and cytokines	(39,52-57,64,65)
Proinflammatory cytokine release	High production of TNF-α, IL-1β, chemokines, leukotrienes, adhesion molecules, ROS and nitric oxide; immature neutrophils show high TNF-α/IL-10 ratio	(62,64)
Immunosuppressive features	Neutrophil subpopulation suppresses T cell proliferation via hydrogen peroxide release	(40)
Mature vs. immature neutrophils	Immature neutrophils exhibit prolonged lifespan, decreased phagocytosis/migration and proinflammatory profile; mature neutrophils exhibit stronger innate immune functions	(64,65)
Aged neutrophils	Decreased return to bone marrow; preferential migration to inflammation sites; enhanced phagocytic activity compared with non-aged neutrophils	(66)

[i] NET, neutrophil extracellular trap; ROS, reactive oxygen species; GRK, G protein-coupled receptor kinase; LPS, lipopolysaccharide; TNF-α, tumor necrosis factor-α.

A UK cohort study identified neutrophil dysfunction in sepsis, including a notable and sustained reduction in NETosis (active NET release accompanied by cell death), along with defective neutrophil migration and delayed apoptosis (51). In addition, Demaret et al (39) found increased neutrophil production and decreased apoptosis in the bone marrow in sepsis. The oxidative burst and phagocytic function of neutrophils in patients with septic shock are increased, but their chemotactic function is strongly inhibited (39,52). Several studies have shown that the migration ability of neutrophils in sepsis patients is reduced (53-56). The underlying mechanism may be that lipopolysaccharide (LPS) and cytokines in sepsis activate G protein-coupled receptor kinase on circulating neutrophils, thereby inducing neutrophils to become desensitized to chemoattractant (57). Martins et al (58) also demonstrated that ROS production is upregulated in neutrophils during sepsis. Notably, neutrophil oxidative burst is decreased in the late stages of sepsis (39). These opposing findings suggested that septic shock may involve a transition from immune activation to immune suppression, during which neutrophils may exhibit immunosuppressive features that compromise the host antimicrobial defense (40,59). For example, Pillay et al (40) identified a neutrophil subpopulation that mediates the suppression of T cell proliferation by releasing neutrophils via local hydrogen peroxide.

The recruitment of bone marrow neutrophils increases and their apoptosis decreases during sepsis. Functional characteristics of neutrophils in patients with septic shock included enhanced respiratory burst and phagocytosis, as well as inhibited chemotaxis. High production of ROS and proinflammatory cytokines by neutrophils at sites far from the initial infection may be part of the pathophysiology of sepsis (56,60,61). Proinflammatory factors, including tumor necrosis factor-α (TNF-α), IL-1β, chemokines, leukotrienes, adhesion molecules, ROS, and nitric oxide, serve an important role in amplifying inflammation, recruiting immune cells and inducing tissue injury during sepsis (62). Decreased motility associated with acute neutrophil activation is hypothesized to play a role in the development of multiorgan failure following sepsis (63).

The functions of mature and immature neutrophils in sepsis are heterogenous. Drifte et al reported that immature circulating neutrophils of patients with septic shock support innate immune defenses to a lesser extent than mature neutrophils (64). Moreover, immature neutrophils possess a longer lifespan and a stronger capacity to resist spontaneous apoptosis (64). Compared with mature granulocytes, the phagocytosis and migration abilities of immature neutrophils are lower (64,65). The high TNF-α/IL-10 ratio in immature neutrophils suggests these cells adopt a proinflammatory profile, characterized by predominant production of inflammatory cytokines (64). Aged neutrophils also have different characteristics. Using a mouse model, Uhl et al (66) reported that the number of aged neutrophils returning to the bone marrow is decreased during an acute inflammatory response to endotoxemia, owing to rapid migration to sites of inflammation. Upon reaching inflamed tissues, aged neutrophils exhibit higher phagocytic activity than subsequently recruited non-aged neutrophils (66).

The abundant phenotypical and functional heterogeneity of neutrophils has been extensively studied, but the underlying mechanisms of diversity remain unclear (21,67). Neutrophils exhibit plasticity, allowing them to respond and adapt to various stimuli. Such context-dependent and reversible changes may resemble stable subsets, thereby confounding the interpretation of true heterogeneity. Because the methodologies used to study neutrophil heterogeneity need improvement and investigations of neutrophil heterogeneity are in an exploratory stage, additional studies are needed to understand whether neutrophil heterogeneity is the result of cell programming or the basic functional activity of cells.

Neutrophil heterogeneity in sepsis by omics

The intricate nature of sepsis and its impact on neutrophil function has prompted researchers to investigate the molecular underpinnings of neutrophil heterogeneity using omics technologies (68-76). Each omics layer sheds light on neutrophil behavior in sepsis, providing insight into their roles in both host defense and immune dysregulation. Table II provides a brief summary of omics techniques (68-76).

Table II Omics levels, research scope and technology.

Table II

Omics levels, research scope and technology.

Omics level	Definition	Research scope	Technology/platform	Explanation	(Refs.)
Genomics	Comprehensive analysis of the genome, including DNA sequence, structural variants and mutations	WGS; WES	Short-read seq (Illumina, Inc.); long-read seq (PacBio, ONT)	WGS covers the entire genome to detect variants; WES focuses on protein-coding regions; long-read seq resolves structural variants and repetitive elements	(68-70)
Epigenomics	Study of genome-wide epigenetic modifications regulating gene expression without altering DNA sequence	DNA methylation profiling; chromatin accessibility mapping; protein-DNA interaction profiling	Bisulfite seq; ATAC-seq; ChIP-seq	DNA methylation profiling reveals CpG methylation; ATAC-seq measures chromatin accessibility; ChIP-seq identifies histone modifications or TF binding	(71)
Transcriptomics	Profiling of RNA expression levels to capture gene activity	Bulk RNA-seq; scRNA-seq; spatial transcriptomics	Short-read sequencing (Illumina, Inc.); in situ capture platforms (10x Genomics, Slide-seq)	Bulk RNA-seq quantifies average transcript abundance; scRNA-seq resolves cell-to-cell heterogeneity; spatial transcriptomics retains tissue architecture while measuring gene expression	(72)
Proteomics	Global profiling of protein abundance, modification and interactions	Protein identification and quantification	MS-based proteomics (LC-MS/MS, DIA, TMT); antibody-based proteomics (RPPA, CyTOF)	MS-based methods enable high-throughput protein identification and quantification; antibody-based methods provide sensitive detection of specific proteins	(73,74)
Metabolomics	Systematic identification and quantification of metabolites reflecting cell metabolic state	Targeted and untargeted metabolomics	LC-MS, GC-MS, NMR platforms	Targeted metabolomics measures predefined metabolites with high sensitivity; untargeted metabolomics broadly surveys metabolites to capture global metabolic changes	(75,76)

[i] WGS, whole-genome sequencing; WES, whole-exome sequencing; ONT, Oxford Nanopore Technologies plc; PacBio, Pacific Biosciences; ATAC, assay for transposase-accessible chromatin; seq, sequencing; ChIP, chromatin immunoprecipitation; sc, single-cell; LC-MS/MS, liquid chromatography tandem mass spectrometry; DIA, data-independent acquisition; TMT, tandem mass tags; RPPA, reverse phase protein array; CyTOF, cytometry by time of flight; GC, gas chromatography; NMR, nuclear magnetic resonance.

Genome

Genomic studies have revealed polymorphisms and mutations within genes associated with neutrophil functions that may predispose individuals to sepsis or influence disease severity (77-80). Andiappan et al (77) investigated the expression quantitative trait loci (eQTL) of neutrophils and identified 21,210 eQTLs on 832 unique genes. The aforementioned study used Ingenuity Pathway Analysis to reveal an enrichment of neutrophil eQTLs in inflammatory disease, consistent with the established role of neutrophils in host defense against pathogens (77). Utilizing a genome-wide association study, Wang et al (78) investigated NET biomarkers to explore the causal association between NET and sepsis. Myeloperoxidase (MPO)-DNA complex is a biomarker of NET (79). With every standard deviation increase in the levels of the MPO-DNA complex, there is an ~18% increase in the risk of sepsis, a 51% increase in the risk of 28-day death from sepsis, ~38% rise in the risk of requiring intensive care due to sepsis and ~125% higher risk of 28-day death from sepsis requiring intensive care (80). Elevated NET levels may elevate the risk of sepsis onset, progression and mortality (80).

Beyond the genome, the epigenome provides an additional layer of regulation that shapes neutrophil heterogeneity. Epigenetic mechanisms such as histone modification, DNA methylation and chromatin accessibility dynamically modulate transcriptional programs, enabling neutrophils to adopt diverse functional states in response to microenvironmental cues. Using chromatin immunoprecipitation sequencing of histone H3K4me3, Piatek et al (80) characterized the epigenetic regulation of human neutrophil plasticity and heterogeneity following stimulation with LPS, TNF-α or IL-10. The aforementioned study revealed that changes in H3K4me3-marked transcriptional start sites are associated with diverse functional programs, including neutrophil activation, cytokine production, apoptosis, histone remodeling and NF-κB signaling pathways (80). IL-10 induces a distinct subset of apoptotic yet transcriptionally active neutrophils, which display a non-canonical NF-κB driven cytokine profile while simultaneously suppressing the canonical NF-κB pathway (80). These findings highlight that epigenomic profiling uncovers previously unrecognized heterogeneity in neutrophil functional states, and that H3K4me3-associated DNA binding sites may serve as potential therapeutic targets for immunomodulation.

Transcriptome

Building on the genomic foundation, transcriptomic studies have advanced understanding of neutrophil heterogeneity (81-88).

Bulk transcriptomic profiling in patients with sepsis has identified 37 differentially expressed genes, with functional enrichment and protein-protein interaction analyses highlighting immune and inflammatory signaling pathways, including the PI3K/AKT axis, as key regulators of neutrophil specialization (81). Validation in whole-blood neutrophil samples from patients with sepsis patients further confirmed these transcriptional alterations (81). These findings demonstrate sepsis-associated transcriptional alterations, but the limited resolution of bulk transcriptomics obscures cellular heterogeneity and prevents the identification of distinct neutrophil subsets. scRNA-seq provides a deeper resolution. Xie et al (28) identified eight transcriptional neutrophil clusters in mice, with G0-G4 representing bone marrow developmental stages and G5a-c representing three transcriptionally distinct mature neutrophil subsets in peripheral blood. Bacterial infection accelerates the transition from immature to mature states without altering overall heterogeneity but induces stage-specific transcriptional reprogramming: Early progenitors upregulate genes regulating immune effector functions and ROS metabolism, while mature subsets exhibit increased cytokine production (28). Infection also reshapes transcription factor networks, with defense-associated factors such as interferon regulatory factor 7 activated in immature cells and metabolic regulators such as forkhead box protein P1 (Foxp1) and CCCTC-binding factor (Ctcf) downregulated in mature subsets, suggesting a redistribution of cell resources toward host defense (28). However, a murine model raises questions about their direct applicability to human sepsis, where immune and inflammatory contexts may differ (28). In human sepsis, Xu et al (82) identified novel neutrophil subtypes enriched in late differentiation stages, characterized by upregulation of alkaline phosphatase, liver/bone/kidney (ALPL), CD177 molecule (CD177), S100 calcium-binding protein A8 (S100A8), S100A9), and syntaxin binding protein 2 (STXBP2), providing potential biomarkers for therapy. Hong et al (83) further categorized neutrophils into four subsets (Neu1-Neu4) by single-cell transcriptomic profiling, with Neu1 expansion associated with septic shock and higher Sequential Organ Failure Assessment (SOFA) scores. A Neu1-specific gene module, including NFKBIA (NFKB inhibitor alpha, an inhibitor of NF-κB signaling), CXCL8 (C-X-C motif chemokine ligand 8, encoding interleukin-8), G0S2 (G0/G1 switch gene 2, a regulator of cell cycle and apoptosis), and FTH1 (ferritin heavy chain 1, involved in iron storage and oxidative stress response), demonstrates strong predictive value for shock. Consistently, Neu1 expresses cell surface markers such as CD123, CD38 and CD69, which have been linked to poor prognosis (84-87). However, whether these subsets represent stable cell states or transient activation phenotypes remains unresolved.

Beyond proinflammatory subtypes, immunosuppressive phenotypes have also been reported: Qi et al (88) identified a PD-L1high neutrophil subset induced via the p38α pathway, involving mitogen- and stress-activated kinase 1 (MSK1) and MAPK-activated protein kinase 2 (MK2),, capable of suppressing T cell activation and promoting apoptosis or trans-differentiation (88).

Collectively, transcriptomics reveals that neutrophil heterogeneity in sepsis arises from dynamic transcriptional reprogramming, encompassing both proinflammatory and immunosuppressive subsets with distinct clinical relevance.

Proteome

Transcriptomic profiling has advanced understanding of neutrophil gene expression patterns in sepsis, but does not directly link transcriptional changes to protein function. Proteomics fills this gap by quantifying protein abundance and activity, thereby capturing more immediate functional adaptations of neutrophils.

Tak et al (89) identified CD62Ldim neutrophils as a distinct subset that typically resides outside circulation but is recruited during acute inflammation. Their proteomic signature, enriched in proteins associated with adhesion, activation and immune regulation, underscores functional specialization beyond conventional morphological classification (89). However, whether these proteomic differences translate into stable functional phenotypes or reflect transient activation states remains uncertain, especially since CD62L shedding can occur in multiple inflammatory contexts (89). This raises the question of whether CD62Ldim neutrophils represent a subset or activation-driven state detectable by proteomics but less evident in physiological conditions. In parallel, hypoxia, a hallmark of sepsis pathophysiology, triggers notable proteome remodeling (90). Watts et al demonstrated that hypoxic neutrophils upregulate inflammatory receptors, including formyl peptide receptor (FPR) and granulocyte-macrophage colony-stimulating factor receptor beta chain (GM-CSF receptor β), enhance lysosomal protein scavenging and sustain biosynthesis of granule and cytoskeletal proteins through de novo synthesis (91). These findings highlight the metabolic adaptability of neutrophils under stress (91). The reliance on murine models and controlled hypoxic exposure limits direct extrapolation to human sepsis, where hypoxia is heterogeneous, dynamic and often accompanied by additional insults such as acidosis or oxidative stress (90,91). Furthermore, whether such proteomic adaptations enhance host defense or contribute to maladaptive inflammation remains unresolved.

Taken together, proteomic profiling demonstrates that neutrophils are not passive effectors but metabolically flexible cells capable of remodeling the proteome in response to inflammatory and metabolic stressors. Nevertheless, studies are largely descriptive and rely on either ex vivo stimulation or animal models, which may not capture the temporal and spatial complexity of sepsis in patients (89-91).

Metabolome

Proteomics reveals alterations in signaling molecules, surface receptors and effector proteins that directly regulate neutrophil behavior but does not capture the intricate biochemical pathways that govern these changes in cell function and energy metabolism. Metabolic reprogramming refers to the dynamic reshaping of cell metabolic pathways in response to environmental and functional demands (92). In neutrophils, metabolomics has revealed that this process extends beyond glycolysis to include oxidative phosphorylation and the pentose phosphate pathway, enabling distinct effector functions such as chemotaxis, ROS generation and NET formation (93). This metabolic adaptability constitutes a fundamental basis of neutrophil heterogeneity, particularly in sepsis.

In a recent study, Li et al (94) observed significant alterations in neutrophil metabolism as severe corona virus disease 2019 (COVID-19) progresses, particularly in amino acid, redox and central carbon metabolism. Metabolic changes in neutrophils are associated with decreased activity of the glycolytic enzyme GAPDH (94). When GAPDH is inhibited, glycolysis is suppressed, leading to an increase in pentose phosphate pathway activity, but this also diminishes the neutrophil respiratory burst (94). Furthermore, inhibiting GAPDH triggers the formation of NETs, which depends on the activity of neutrophil elastase (94). This inhibition results in an increase in neutrophil pH and blocking this pH rise prevents both cell death and NET formation (94). These findings indicated that neutrophils in severe COVID-19 exhibit a heterogeneous and dysfunctional metabolic profile, which contributes to their impaired function.

Multi-omics to explore neutrophil heterogeneity in sepsis

Each layer of omics offers insight into genomic variants, transcriptional changes, protein modification or metabolic shift. However, they often present a limited view of the complex interplay between these biological layers. For example, genomics identifies genetic predispositions that shape neutrophil responses, but does not capture the dynamic processes that occur during gene expression and protein translation. Similarly, transcriptomics reveals altered gene expression profiles associated with sepsis but may overlook the functional implications of changes at the protein level. Transitioning from single to multi-omics allows for a more holistic perspective by integrating data across different layers of biological information. Multi-omics approaches not only facilitate the identification of distinct neutrophil subsets and their functional states in real time but also enhance understanding of how metabolic changes influence neutrophil activity during sepsis (Fig. 2).

Figure 2 Multi-omics approaches reveal neutrophil heterogeneity and clinical implications in sepsis. Omics layers include genome, transcriptome, proteome, metabolome and microbiome. Neutrophil subsets and characteristics are identified through single-omics studies and multi-omics integration in sepsis. Potential clinical applications of neutrophil heterogeneity profiling include early detection, patient stratification, targeted therapy, prognostic modeling and precision medicine. eQTL, expression quantitative trait loci; MPO, myeloperoxidase; H3K4me3, histone H3 lysine 4 trimethylation; PI3K/AKT, phosphatidylinositol 3-kinase/protein kinase B; PD-L1, programmed death-ligand 1; MKI67, marker of proliferation Ki-67; CYP1B1, cytochrome P450 family 1 subfamily B member 1; CEACAM8, carcinoembryonic antigen-related cell adhesion molecule 8; S100A8/9, S100 calcium-binding protein A8/9; LDHA, lactate dehydrogenase A; Neu1, neutrophil subtype 1.

Transcriptome and proteome

Integration of transcriptomic and proteomic data has advanced understanding of neutrophil heterogeneity and its functional consequences during sepsis. By linking gene expression with protein abundance and post-translational modification, these approaches not only delineate the molecular programs of neutrophils but also capture how these programs are dynamically reconfigured under septic conditions.

The source of neutrophil heterogeneity remains debated. On one hand, studies have suggested that circulating neutrophils follow a predefined differentiation trajectory derived from hematopoietic progenitor cells (95,96). On the other hand, in vitro experiments have shown that stimulation of mature neutrophils with inflammation-associated molecules can also reshape their transcriptome, indicating that microenvironmental cues serve a crucial role in driving functional diversity (88,97). Thus, both developmental origin and peripheral reprogramming may contribute to heterogeneity. Kaiser et al (98) used transcriptomics and proteomics to reveal mechanisms associated with functional reprogramming of peripheral neutrophils during acute infection. Transcriptome analysis revealed increased expression of classical neutrophil markers such as CXCL8, SOD2, S100A8/A9, CSF3 receptor and myeloid cell nuclear differentiation antigen (MNDA) (98). Activation markers and antimicrobial genes are upregulated, including cystatin F (CST7), S100 calcium binding protein A12 (S100A12), interleukin 1 receptor type 2 (IL1R2) and annexin A1 (ANXA1) (98). The aforementioned study used mass spectrometry to assess whether post-infection transcriptomic alterations are reflected in the proteome of human neutrophils (98). Corresponding upregulation at both transcriptional and protein levels included interferon-induced transmembrane protein 3 and alkaline phosphatase (98). In addition, patients with acute bacterial infections show enrichment of CD177high neutrophils (98). Both transcription and protein expression of CD177 are positively associated with disease severity (98). These results suggest that the transcriptome response of neutrophils is effectively translated into the proteome during bacterial inflammation. However, the aforementioned study did not resolve how transcriptomic shifts intersect with upstream regulatory mechanisms such as chromatin accessibility and transcription factor activity (95), leaving the drivers of neutrophil reprogramming incompletely understood. Moreover, although CD177 expression is associated with disease severity, its biological role remains unclear, especially since up to 10% of individuals lack CD177 without apparent immune defects (99-101). This raises the possibility that CD177 serves as a context-dependent marker rather than a direct effector. Finally, the cohort size and baseline heterogeneity may limit generalizability, and larger longitudinal studies are required to clarify the stability and functional relevance of CD177+ neutrophils in bacterial infection (99-101).

In parallel, Kwok et al (102) identified a programmed neutrophil lineage within granulocyte-monocyte progenitors (GMPs) and analyzed the heterogeneous neutrophil subsets at different stages of this lineage in sepsis. During the differentiation of GMPs into preNeus, the aforementioned study described two phenotypically distinct types of neutrophil progenitors, termed proNeus (102). These include the CD34hi CD106- CD11blo proNeu1 and the CD34lo CD106+ CD11bhi proNeu2 subset (102). Proteomics shows progressive upregulation of CD11b and downregulation of CD34 expression during the differentiation of GMP into preNeus (102). Other differentially expressed cell surface markers include CD81, CD49a, CD106 and CD63, which may serve as positive or exclusive markers (102). Transcriptomics confirms the downregulated expression of lineage-associated genes and upregulated expression of granule protein genes in the GMP differentiation trajectory during sepsis, suggesting that sepsis promotes ProNeu1 differentiation (102). ProNeu1 and proNeu2 are both early neutrophil progenitor cell populations, but their functions are not identical (102). ProNeu1 exhibits a stronger capacity for proliferation than proNeu2 (102). During the early stages of sepsis inflammation, proNeu1 expands specifically and extensively, which decreases monocyte differentiation (102). However, proNeu2 remains largely unchanged during infection (102).

A recent multi-omics study by Kwok et al demonstrated that septic neutrophils acquire immunosuppressive properties and are enriched in patients with the sepsis response signature group 1 (SRS1) (103). Single-cell transcriptomics and cell surface protein profiling reveal the expansion of immature and functionally distinct neutrophil subsets, including CEACAM8+ degranulating cells, S100A8/9hi cells, IL1R2+, peptidyl arginine deiminase type 4+, MPO+ and proliferative MK167+ CYP1B1+ neutrophils, many of which are specific to sepsis rather than sterile inflammation (103). Coculture assays further demonstrate that these septic subsets inhibit CD4+ T cell activation (103). Epigenomic profiling of hematopoietic stem and progenitor cells revealed CCAAT/enhancer-binding protein α (CEBPA)- and β (CEBPB)-driven regulatory programs consistent with the activation of both steady-state and emergency granulopoiesis, linking neutrophil reprogramming to systemic alteration in hematopoiesis (103). Importantly, within the SRS1 subtype, neutrophil subsets such as IL1R2+ and MK167+ CYP1B1+ cells are significantly enriched and exhibit STAT3- and CEBPB-dependent gene expression programs, indicating that neutrophil heterogeneity is not only a reflection of sepsis pathology but also a driver of immunosuppressive disease endotypes (103). However, the single-cell cohorts analyzed may not represent the spectrum of sepsis, and causal involvement of transcription factors such as CEBPB and STAT3 requires direct experimental validation.

Transcriptome and metabolome

Integration of transcriptomic and metabolomic approaches provides insights into how metabolic reprogramming shapes neutrophil heterogeneity during sepsis.

Neutrophils are reliant on glycolysis to fuel key antimicrobial functions, including chemotaxis, phagocytosis, oxidative burst and NET formation (104-106), yet their metabolic flexibility is altered in the septic milieu. Pan et al (107) demonstrated that sepsis-tolerant neutrophils exhibit reduced glycolytic activity associated with downregulation of LDHA via the PI3K/Akt/hypoxia-inducible factor (HIF)-1α pathway, leading to impaired chemotactic and phagocytic capacity. Metabolomic profiling further revealed that lactate levels are diminished in septic neutrophils compared with non-septic infected controls, highlighting a distinct metabolic state of glycolytic suppression (107).

Lactate, long regarded as a metabolic byproduct, is a immunomodulatory metabolite capable of exerting feedback control over immune cell metabolism (108,109). While evidence in monocytes and macrophages has demonstrated lactate-driven immunosuppression (110,111), its role in neutrophil biology remains less well understood. Pan et al (107) suggested that reduced lactate production in sepsis may represent a unique metabolic signature of neutrophil dysfunction, warranting further exploration of lactate-mediated feedback in neutrophil plasticity. At the molecular level, transcriptomic changes in key glycolytic enzymes, including LDHA, pyruvate dehydrogenase kinase 1, glucose transporter 1 (GLUT1) and pyruvate kinase M2 (PKM2) converge with metabolomic alterations, reinforcing the key role of glycolysis in neutrophil effector responses (107). Stabilization of HIF-1α restores LDHA expression and glycolytic activity, thereby rescuing neutrophil chemotaxis and phagocytosis, underscoring the PI3K/Akt/HIF-1α axis as a central regulator of neutrophil function in sepsis (107).

Collectively, combined transcriptomic and metabolomic analyses uncover a metabolically defined neutrophil subpopulation in sepsis characterized by glycolytic suppression and functional impairment (107). This multi-omics perspective not only advances understanding of neutrophil heterogeneity but also highlights metabolic checkpoints such as glycolysis and HIF-1α signaling as potential therapeutic targets to restore neutrophil immunity in sepsis (107).

Proteome and metabolome

Exploring the association between protein expression profiles and metabolic signatures reveals key regulatory hubs and pathways that influence neutrophil activation, survival and effector functions.

Using a proteomic approach, Parthasarathy et al (112) examined the differential expression of neutrophil subsets in patients with sepsis. The expression of CD10, CD16 and CD86 is downregulated, while human leukocyte antigen- DR isotype (HLA-DR), CD11b, CD80, CD184, CD63 and CD66b are upregulated in sepsis (39,112,113). The aforementioned study identified a specific mature neutrophil subset with high expression of CD274 (PD-L1) and CD300f. Previous studies have shown that blocking of PD-L1 improves survival in patients with sepsis (114-116), and CD300f deletion stimulates neutrophil recruitment to the site of infection and decreases septic death in mice (117). According to the expression of CD177, immature neutrophil subsets are divided into two subgroups: CD10- CD177+ and CD10- CD177- (113). CD177 rapidly mobilizes specific particles to the cell surface following cell activation (112). Differential expression of CD184 (CXCR4) and HLA-DR in CD10- CD177+ immature neutrophil subsets is observed in patients with sepsis (118). Aged neutrophils expressing CD184 exhibit a higher migratory activity and phagocytic capacity than the subsequently recruited non-aged neutrophils (65). HLA-DR expression was increased in CD10- CD177+ septic neutrophils (112). The expression and role of HLA-DR on neutrophils remains unclear (119,120). Soluble factors pentraxin 3 (PTX3), angiopoietin-2 (Ang-2), endothelial cell-specific molecule-1 (Endocan), growth arrest-specific 6 (Gas6) and the inflammatory marker procalcitonin are upregulated in sepsis (112). These factors are elevated in patients with sepsis and contribute to vascular leak and endothelial dysfunction (121,122). Correlation analysis shows that CD10 is inversely correlated with these factors (112). Notably, immature neutrophils store and release PTX3 during inflammation, and this factor predicts disease severity and mortality in sepsis (123-126). Immature neutrophils may be a driver of vascular inflammation or leak in sepsis.

While these observations provide insights into the functional diversity of neutrophil subsets and their potential links to soluble mediators and vascular pathology, limitations remain. Distinguishing sepsis from other infections or organ failure remains clinically challenging. Integrated multi-omics approaches have revealed associations between neutrophil phenotypes, soluble mediators and metabolic alterations. However, these findings are largely observational and derived from limited patient cohorts (112). Therefore, further mechanistic studies are needed to validate whether specific neutrophil subsets and their products directly drive vascular leakage and immune dysregulation in sepsis.

Proteome and microbiome

The combined investigation of proteome and microbiome has revealed the functional dynamics of neutrophils in the context of sepsis.

Wang et al (127) isolated neutrophils from patients with sepsis after surgery and characterized intracellular bacterial communities, also termed as the neutrophil-specific microbiome. The aforementioned study showed that the proportion of actinobacteria decreased, while the levels of proteobacteria increased (127). Compared with healthy controls, the abundance of Escherichia/Shigella, Klebsiella and Bradyrhizobium is higher in patients with sepsis (127). The neutrophil-specific microbiome of patients with sepsis exhibits heterogeneity (127). Dysregulation of the circulating microbiota may increase the risk of postoperative infectious events (128). In addition, quantitative proteomic analysis of neutrophils derived from patients with demonstrates proteins involved in bactericidal activities of neutrophils are downregulated, especially in patients with septic shock (127). Significant downregulation of some immunomodulatory-associated proteins is also observed in sepsis patients, including integrin α-M, IgA Fc receptor and lactotransferrin (127). MMP9 is significantly downregulated in patients with septic shock, indicating that the migratory activity of neutrophils is impaired (128). Proteomic analysis reveals a decrease in neutrophil function in sepsis (127).

Beyond elucidating the molecular heterogeneity of neutrophils, multi-omics suggests potential biomarkers and therapeutic targets with translational relevance. Table III summarizes representative neutrophil-associated biomarkers identified from multi-omics, along with their potential diagnostic or prognostic value in sepsis (28,77,78,80-83,88,89,91, 94,98,102,103,107,112,115,117,127).

Table III Clinical translation and biomarker potential of neutrophil signatures identified by multi-omics in sepsis.

Table III

Clinical translation and biomarker potential of neutrophil signatures identified by multi-omics in sepsis.

Biomarker/subtype	Omics source	Clinical relevance	(Refs.)
Neutrophil eQTL	Genomics	Identification of 21,210 neutrophil eQTLs (832 genes) underscores the role of neutrophils in host defense and inflammation	(77)
MPO-DNA complex	Genomics	Elevated MPO-DNA associated with increased risk of sepsis (18%), 28-day mortality (51%) and ICU requirement (38%) and mortality (125%)	(78)
H3K4me3-marked subsets	Epigenome	IL-10 induces transcriptionally active neutrophils, contributing to inflammation resolution; H3K4me3 positioning suggests potential immunotherapeutic targets	(80)
Transcriptional neutrophil clusters	Transcriptome (scRNA-seq)	Infection accelerates immature-to-mature transition; developmental stage-dependent neutrophil responses to infection	(28)
DEGs	Transcriptome (bulk RNA-seq)	DEGs enriched in immune/inflammatory signaling and PI3K/AKT pathway; AKT1 validated as a potential therapeutic target in sepsis	(81)
Neutrophil subtypes enriched at late differentiation stage	Transcriptome (scRNA-seq)	A total of two novel subsets enriched in human sepsis; five hub genes validated in mice; candidate biomarkers and therapeutic targets (ALPL, CD177, S100A8/A9, STXBP2)	(82)
Neu1	Transcriptome (scRNA-seq)	Neu1 associated with septic shock and SOFA score; Neu1_C module predictive for shock (AUC=0.81); expresses poor prognosis markers (CD123, CD38, CD69)	(83)
PD-L1high neutrophils	Transcriptome (scRNA-seq)	Immunosuppressive subset in sepsis (p38α/MSK1/MK2 dependent); directly inhibits T cell response; associated with immune suppression and poor host defense; potential therapeutic target via PD-L1 blockade	(88)
CD62Ldim neutrophils	Proteome	Distinct subset recruited during acute inflammation; proteomic profile enriched in adhesion/activation pathways; potential biomarker of systemic inflammation	(89)
Hypoxia-induced neutrophil proteomic remodeling	Proteome	Inflammatory receptor upregulation (FPR, GM-CSFRβ) and lysosomal protein scavenging sustain neutrophil metabolism under stress; metabolic pathways as potential therapeutic targets in hypoxic inflamed tissue	(91)
GAPDH-inhibited neutrophils in severe COVID-19	Metabolome	Dysregulated amino acid, redox and carbon metabolism; suppressed glycolysis with PPP shift; GAPDH inhibition triggers pathogenic NET formation; links neutrophil metabolic dysfunction to tissue damage and highlights GAPDH as a potential therapeutic target	(94)
CD177high neutrophils	Transcriptome and proteome	Expanded during acute bacterial infection; TLR4/NF-κB driven transcriptional plasticity enhances antibacterial programs; CD177 upregulation is associated with inflammation severity and may serve as a marker of acute infectious states	(98)
CD34hi CD106- CD11blo ProNeu1	Transcriptome and proteome	Early committed progenitor; expands in early sepsis with high proliferative capacity, suppressing monocyte output; potential biomarker of emergency granulopoiesis	(102)
CD34lo CD106+ CD11bhi ProNeu2	Transcriptome and proteome	Intermediate progenitor; stable during sepsis; reference subset for maturation trajectory	(102)
IL1R2+, MK167+ CYP1B1+, CEACAM8+, S100A8/9hi	Transcriptome (scRNA-seq) and proteome	Sepsis-specific; inhibit CD4+ T cells; enriched in SRS1 endotype; associated with STAT3/CEBPB-driven immunosuppression	(103)
LDHAlow glycolytic neutrophils	Transcriptome and metabolome	Exhibit suppressed glycolysis with impaired chemotaxis and phagocytosis; dysfunction mediated by PI3K-Akt-HIF-1α-LDHA axis; reversible by HIF-1α activation, suggesting therapeutic potential for restoring immune competence in sepsis	(107)
CD10- CD177+ immature neutrophils	Proteome and metabolome	Associated with SOFA score and endothelial dysfunction (PTX3, Ang-2, endocan); PTX3 release associated with severity and mortality; associated with metabolic dysregulation and identified as a feature for patient stratification	(112)
PD-L1high neutrophils	Proteome and metabolome	PD-L1 blockade improves survival in murine CLP sepsis model	(115)
CD300f+ neutrophils	Proteome and metabolome	Disruption or deletion of CD300f decreases septic mortality in murine CLP model of sepsis	(117)
Neutrophil microbiome shifts	Proteome and microbiome	Associated with impaired bactericidal activity and migration; downregulation of ITGAM, LTF and MMP9; associated with septic shock	(127)

[i] eQTL, expression quantitative trait loci; MPO, myeloperoxidase; DEG, differentially expressed gene; seq, sequencing; sc, single-cell; SOFA, Sequential Organ Failure Assessment; PD-L1, programmed death ligand 1; CD62L, CD62 ligand; LPS, lipopolysaccharide; TNF-α, tumor necrosis factor α; ChIP, chromatin immunoprecipitation; H3K4me3, histone H3 lysine 4 trimethylation; PI3K, phosphoinositide-3-kinase; STXBP2, syntaxin-binding protein 2; MK2, MAPK-activated protein kinase 2; FPR, formyl peptide receptor; GM-CSFR, granulocyte-macrophage colony-stimulating factor receptor; PPP, pentose phosphate pathway; TLR4, toll-like receptor 4; C/EBP, CCAAT/enhancer-binding protein; LDHA, lactate dehydrogenase A; HIF-1α, hypoxia-inducible factor 1 α; PTX3, pentraxin 3; Ang-2, angiopoietin-2; endocan, endothelial cell-specific molecule-1; CLP, cecal ligation and puncture; ITGAM, integrin α M; LTF, lactotransferrin.

Challenges and limitations

Despite the advances in understanding neutrophil heterogeneity in sepsis through multi-omics, key challenges hinder the full potential of these approaches.

Patient heterogeneity is a key obstacle to characterizing neutrophil heterogeneity in sepsis through multi-omics approaches. Variations in demographic and physiological backgrounds (age, sex, genetic factors) shape neutrophil development and lifespan, potentially masking sepsis-specific changes (129-132). Underlying comorbid conditions may introduce baseline alterations in neutrophil function, thereby contributing to heterogeneity and complicating the attribution of omics signatures solely to sepsis (133). In addition, pathogen type, primary infection site and therapeutic intervention may induce distinct neutrophil activation programs, while clinical trajectories range from hyperinflammation to immunosuppression, generating variable molecular patterns (33,134-136). These sources of variability make datasets harder to compare, decrease reproducibility and blur the true sepsis-associated neutrophil signatures. To overcome this challenge, future studies should incorporate strategies such as stratified analyses, matched cohort designs and standardized metadata collection to distinguish patient-level variability from the intrinsic biological diversity of neutrophils in sepsis.

Sample preparation is a key step in multi-omics studies, but introduces inherent biases, particularly during cell isolation and processing. For example, methods based on density gradient centrifugation may inadvertently exclude low-density neutrophils (LDNs), which are typically found in the peripheral blood mononuclear cell fraction (137). By contrast, commonly used methods obtain samples from peripheral blood or tissue, followed by red blood cell lysis or enzymatic digestion (138). The cells are centrifuged, resuspended, washed and filtered to ensure high-quality viable cells. Neutrophils are purified by either positive or negative selection using magnetic beads or fluorescence-activated cell sorting and subjected to scRNA-seq for downstream analysis (28). Other improvements include initial selection of total neutrophils using magnetic beads followed by density gradient separation of LDNs (139,140). In addition, density gradient-based approaches to neutrophil isolation may result in preparations containing small numbers of contaminating leukocytes, primarily eosinophils, a potential source of bias despite the small contribution of these leukocytes to the overall gene expression profile (141). These technical variations can lead to discrepancies between in vitro or ex vivo findings and the actual in vivo behavior of neutrophils, particularly within the complex microenvironment of sepsis.

The temporal dynamics of neutrophil changes during sepsis are not well understood. The heterogeneity of neutrophils in sepsis results from a complex interaction between intrinsic factors and disease-associated changes over time. Sepsis is a dynamic condition, marked by rapid shifts in immune response, from excessive inflammation in the early stage to immunosuppression in the later stages (33). However, most current multi-omics studies focus on single time-point analyses, providing a static snapshot of neutrophil phenotypes, which may overlook the ongoing changes in neutrophils during the immune transition in sepsis (83,112). Therefore, longitudinal multi-omics studies are necessary to capture the immune switching of neutrophils over time and improve understanding of this dynamic process.

Each omics layer provides a unique resolution and focus for capturing biological heterogeneity, yet these differences also impose analytical limitations. The integration of multi-omics data from high-throughput platforms inherently faces challenges due to their diverse characteristics (142). Picard et al (143) highlighted that the heterogeneity of data complicates the integration process. For example, transcriptomic data is often subjected to RNA-seq normalization, while proteomic data typically rely on mass spectrometry-specific scaling methods, leading to differing data ranges and distribution patterns. These disparities necessitate alignment procedures before integration. Data quality also poses a concern. Issues such as noise, missing values and batch effects notably impact the outcomes of the analysis (144). In addition, there are kinetic differences between RNA and protein expression. For example, Hoogendijk et al (145) reported that nearly 30% of the transcriptome -proteome pairs showed inconsistent dynamics during neutrophil differentiation. This discrepancy may contribute to the inconsistency between the transcriptome data from scRNA-seq and the protein-based surface marker data from mass cytometry. These mismatches complicate the classification of cell populations and the annotation of functional roles.

Conclusion

Sepsis is a life-threatening syndrome characterized by a systemic inflammatory response to infection, leading to multiorgan dysfunction and potential mortality. The pathophysiology of sepsis involves complex interactions between pathogens, the immune system and various host factors. Advancements in research have demonstrated the key role of immune cells in sepsis, particularly neutrophils, which are frontline responders in the immune system. The functional and phenotypical heterogeneity of these immune cells during sepsis can influence outcomes.

In terms of phenotypical diversity, neutrophils in sepsis exhibit various surface markers and morphologies that reflect their activation state, origin and roles in the inflammatory response. Functionally, neutrophils exhibit heterogeneity in their ability to degranulate, phagocytose, release ROS, form NETs, migrate, undergo apoptosis and mediate immunosuppression. Understanding these functional differences is key for developing targeted therapeutic strategies aimed at modulating neutrophil activity during sepsis.

The study of neutrophil heterogeneity in sepsis through multi-omics approaches has provided insights into the complex and adaptive nature of these immune cells. Multi-omics studies have revealed that the differentially expressed genes, proteins and metabolites underlying neutrophil heterogeneity are not independent events but reflect interconnected layers of regulation. Transcriptomic analyses have identified altered expression of transcription factors and signaling molecules that reprogram neutrophil activation, survival and differentiation, while proteomic profiling has identified changes in effector functions, including degranulation, phagocytosis and cytokine release (81,88). Metabolomic data have further indicated shifts in glycolysis, oxidative phosphorylation and amino acid metabolism that provide energetic and biosynthetic support for these functions. In addition, epigenomic profiling has elucidated associations between chromatin accessibility and histone modifications with NF-κB signaling, apoptosis and cytokine regulation, thereby uncovering previously unrecognized neutrophil subsets (80,107). Multi-omics integration has identified transcription factor-driven programs, such as CEBPA/CEBPB- and STAT3-dependent networks, which connect emergency granulopoiesis with the emergence of sepsis-specific neutrophil subsets (103). Collectively, these interconnected regulatory networks provide a mechanistic basis by which neutrophils acquire distinct functional states, thereby contributing to the phenotypical and functional heterogeneity observed in sepsis.

Availability of data and materials

Not applicable.

Authors' contributions

ZT designed the study. DC, PZ, JL, SC, SG, YC, YS, TT, LD and TC performed the literature review. ZL wrote the manuscript. CZ edited the manuscript. All authors have read and approved the final manuscript. Data authentication is not applicable.

Ethics approval and consent to participate

Not applicable.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Abbreviations:

Acknowledgements

Not applicable.

Funding

The present study was supported by Hubei Province health and family planning scientific research project (grant no. WJ2023M015).

References