Diagnostic Value of Heparin-Binding Protein (HBP) in Pulmonary Infection After Radical Surgery for Lung Cancer: A Retrospective Study.

1
Introduction
Lung cancer is the leading cause of cancer‐related death worldwide. Lobectomy with mediastinal lymph node dissection is the preferred treatment for patients with early‐ and mid‐stage lung cancer [1]. Video‐assisted thoracoscopic lobectomy (VATS), which is less invasive and promotes faster recovery while reducing complications and mortality compared to open lobectomy, has become widely used [2, 3]. However, due to factors like poor sputum clearance after surgery, reduced immunity, and surgical trauma, pulmonary infection remains a common complication. Clinically, various strategies have been implemented to prevent and manage postoperative infections, including thorough preoperative assessment, patient health optimization, and appropriate antibiotic use. Despite these measures, there are limitations in preventing and managing postoperative pulmonary infections. For example, the rise of antibiotic‐resistant bacteria complicates antibiotic selection, while patient factors such as malnutrition, immunosuppression, and chronic diseases can impair wound healing and increase infection risk [4, 5]. Thus, preventing and treating pulmonary infection after radical lung cancer surgery remains a significant challenge in clinical practice.
In clinical practice, diagnosing postoperative pulmonary infection requires a high index of suspicion and the integration of multiple diagnostic modalities. Critical differentiation is required from non‐infectious complications—including atelectasis, pleural effusion, pulmonary edema, pulmonary embolism, and early anastomotic leakage. Etiological investigations (e.g., sputum culture and lower respiratory tract sampling) remain essential for pathogen identification and constitute a diagnostic gold standard for hospital‐acquired pneumonia (HAP). Imaging methods, such as chest x‐rays and computed tomography (CT) scans, may show pulmonary infiltration, consolidation, or effusion, which supports infection diagnosis, though their sensitivity and specificity are limited [6]. Clinical symptomatology and physical signs provide supplementary diagnostic support. Notably, combined analysis of peripheral blood inflammatory biomarkers enables rapid infection assessment, enhances diagnostic precision for postoperative infections, and optimizes patient monitoring. Hailun et al. [7] demonstrated that detecting serum amyloid A (SAA), C‐reactive protein (CRP), and procalcitonin (PCT) simultaneously can quickly indicate respiratory infection in children, aiding in distinguishing bacterial from non‐bacterial infections. Biomarker‐guided algorithms (e.g., SAA/CRP/PCT triage) may expedite infection identification prior to culture results, particularly valuable in postoperative settings where imaging findings are non‐specific. Integrating these complementary methodologies within a stratified diagnostic framework maximizes therapeutic efficacy and patient outcomes.
Heparin‐binding protein (HBP), also known as cationic antimicrobial protein 37 (CAP37), is a 37‐kDa cationic protein pre‐stored in azurophilic granules and secretory vesicles of neutrophils. Upon neutrophil activation, HBP induces inflammation by altering cytoskeletal dynamics to increase vascular endothelial permeability, thereby promoting neutrophil and fibroblast extravasation [8, 9]. The release of HBP from activated neutrophils—triggered directly by bacterial pathogen‐associated molecular patterns [10]—establishes its utility as a specific biomarker for detecting bacterial infections across diverse clinical contexts [11]. Notably, HBP exhibits high sensitivity and specificity in diagnosing sepsis and predicting organ failure [12, 13, 14], with the 2021 International Guidelines for Sepsis Management designating it a “promising novel diagnostic biomarker” [15]. As a novel inflammatory biomarker, its release kinetics precede those of established biomarkers (e.g., CRP, SAA, and PCT), providing a critical window for early clinical intervention [9, 16, 17, 18, 19]. Studies further support HBP's diagnostic utility in diverse conditions, including pancreatitis, central nervous system infections, and urinary tract infections [20, 21, 22, 23]. However, the diagnostic role of plasma HBP in early pulmonary infection following radical lung cancer surgery remains underexplored. Here, we evaluate the additive diagnostic value of plasma HBP combined with established inflammatory markers, including white blood cell (WBC), CRP, neutrophil (NEUT), lymphocyte (LYMN), neutrophil‐to‐lymphocyte ratio (NLR), interleukin‐6 (IL‐6), and interleukin‐8 (IL‐8), for early detection of pulmonary infection following radical lung cancer surgery.
Given the diagnostic value of inflammatory biomarkers in infectious diseases, the combined detection of multiple biomarkers may improve the early diagnostic and prognostic assessment of these conditions. In this study, lung cancer patients who underwent only radical surgery at Shanghai Chest Hospital, affiliated with the Shanghai Jiao Tong University School of Medicine, were selected. The study aims to identify independent predictors for pulmonary infection and assess the diagnostic value of HBP in combination with other biomarkers for pulmonary infection after radical lung cancer surgery.

2
Methods
2.1
Study Design and Participants
This retrospective, single‐center observational cohort study analyzed clinical data from 292 lung cancer patients who underwent radical surgery at Shanghai Chest Hospital, Shanghai Jiao Tong University School of Medicine, from January 2023 to August 2023. A comprehensive review of the patients' clinical data (including symptoms, physical signs, test results, and imaging tests) was conducted using the hospital's electronic medical record system. Patients were categorized into two groups based on the occurrence of pulmonary infection post‐surgery: the pulmonary infection group (n = 31) and the non‐pulmonary infection group (n = 261). Laboratory data, including HBP, WBC, CRP, and IL‐6 levels, were obtained from peripheral venous blood samples collected within 48 h post‐surgery. Inclusion criteria: (1) patients who underwent radical lung resection in our hospital; (2) primary lung cancer confirmed by postoperative pathological biopsy; (3) no prior anti‐tumor treatments, including radiotherapy, chemotherapy, or immunotherapy, before radical lung cancer surgery; and (4) complete medical records available (including demographic characteristics, comprehensive laboratory test results, imaging results and detailed surgical records). Exclusion criteria: (1) acute or chronic infections before surgery; (2) other malignant tumors; (3) autoimmune diseases; and (4) incomplete or lost data. This study was conducted in compliance with the Declaration of Helsinki and was approved by the Ethics Review Committee of Shanghai Chest Hospital, Shanghai Jiao Tong University School of Medicine (IRB, IS24124).

2.2
Laboratory Procedures
Fasting peripheral venous blood samples were collected from patients within 48 h after radical lung cancer surgery. EDTA‐anticoagulated samples for complete blood count analysis (including WBC, NEUT, and LYMN) and CRP measurement required processing within 30 min of collection, with results reported within 1 h. Sodium citrate‐anticoagulated blood samples for HBP testing and clot activator samples for IL‐6 and IL‐8 quantification were processed within 1 h of collection, with results reported within 2 h. Serum and plasma separated by centrifugation at 2300 g for 10 min at room temperature. WBC and CRP levels were measured using an automated blood analyzer (BC‐6800Plus, Mindray, Shenzhen, China). HBP concentrations were determined using a fluorescence quantitative immunochromatographic kit (Joinstar, Hangzhou, Zhejiang Province, China), IL‐6, and IL‐8 levels were measured with a flow cytometric array luminescence analyzer (iMatrix100, Hangzhou, Zhejiang Province, China). All testing procedures strictly followed the reagent kit instructions.

2.3
Criteria for Pulmonary Infection
The diagnosis of postoperative pulmonary infection was made by thoracic surgeons. In line with international guidelines for hospital‐acquired pneumonia [5], a postoperative pulmonary infection was diagnosed if the patient presented with cough, thick sputum, and wet lung rales, along with at least one of the following: (1) fever (≥ 38.0°C); (2) elevated leukocyte count and/or increased neutrophil proportion; (3) inflammatory lung lesions observed in imaging; and (4) positive pathogenic findings, including pathogen isolation from sputum, blood, lower respiratory tract secretions, or pleural fluid in cases with pleural effusion, or pathogenic evidence from immunoserology or histopathology.

2.4
Statistical Analysis
Normally distributed variables were expressed as mean ± standard deviation (mean ± SD) and compared between groups using an independent samples t‐test. Non‐normally distributed variables were presented as median and interquartile range (IQR) and compared between groups using non‐parametric tests (Mann–Whitney U‐test or Fisher's exact test). For categorical data, values were expressed as counts or percentages, with group comparisons conducted using the χ² test. Spearman correlation coefficients among inflammatory factors were calculated. The binary logistic regression and least absolute shrinkage and selection operator (LASSO) were conducted to identify independent predictors associated with pulmonary infection following radical lung cancer surgery. The predictive value of each inflammatory factor, both individually and in combination, for postoperative pulmonary infection was evaluated by plotting ROC curves and calculating the Youden index to identify the optimal cutoff value. SPSS software (version 26.0, IBM SPSS Statistics) and R software (version 3.5.3, R Foundation for Statistical Computing) were applied for all analyses. A two‐sided p value of less than 0.05 was considered statistically significant.

3
Results
3.1
Demographics and Clinical Characteristics
A total of 292 patients meeting the eligibility criteria for radical lung cancer surgery were included in this study. The cohort consisted of 120 males (41.1%) and 172 females (58.9%), with a mean age of 60.98 ± 12.41 years (range of 20–86 years). Thirty‐one (20 males) developed pulmonary infections, with a median diagnosis time of 4 days (IQR: 4–5), while 261 (100 males) did not, resulting in an incidence rate of 10.6%. Comparative analyses of surgical site, tumor grade, and ASA (American Society of Anesthesiologists Physical Status Classification System, ASA) grade showed no statistically significant differences between the groups (p > 0.05). Gender, age, hospital stay duration, operation duration, pathological type, tumor size, and operation type showed statistically significant differences between the two groups (all p values < 0.05). Inflammatory markers, including HBP, WBC, NEUT, NLR, CRP, IL‐6, and IL‐8, were significantly higher in the pulmonary infection group compared to the non‐pulmonary infection group (p < 0.05). Lymphocyte (LYMN) levels were significantly lower in the pulmonary infection group (p < 0.05). Details are provided in Table 1.

3.2
Clinical Characteristics of Pulmonary Infection Group
Among 31 patients with pulmonary infection, 24 (77.42%) were admitted to the ICU, and 7 (22.58%) were admitted to the general ward. Twenty‐four patients (77.42%) experienced fever (≥ 38.0°C). Additionally, 20 patients (64.52%) exhibited symptoms of cough and sputum, while 4 patients (12.9%) experienced dyspnea post‐surgery. A total of 26 patients with pulmonary infections underwent sputum or pleural fluid cultures, with 15 testing positive for pathogens, representing a positivity rate of 57.69%. Details of the pathogen strains are provided in Figure 1. Chest imaging was performed in 28 patients with pulmonary infections, and 21 were diagnosed with pulmonary infections based on imaging, yielding a positivity rate of 75.0%. See Figure 1 for details.

3.3
Determination of the Independent Predictors for Postoperative Pulmonary Infection
A univariate logistic regression analysis was performed to assess the potential independent predictor of pulmonary infections in patients following radical lung cancer surgery. Results indicated that age, gender, tumor size > 5 cm, open lobectomy, operation duration, HBP, WBC, NEUT, NLR, CRP, and IL‐8 levels were significantly associated with the likelihood of postoperative pulmonary infection (refer to Table 2 for the full list of variables). These factors were subsequently included in a multivariate logistic regression model, which confirmed HBP, NLR, and CRP as independent predictors for pulmonary infection in patients undergoing radical lung cancer surgery (ORs with 95%CI of HBP, NLR, and CRP were: 1.041, 1.024–1.059; 1.132, 1.074–1.193; and 1.05, 1.028–1.073, respectively). See Table 2 for details.
The LASSO regression model was further applied to verify the independent predictors identified by the logistic regression model. The model considered 13 variables, including age, gender, tumor grade, tumor size, operation duration, operation type, HBP, WBC, NEUT, NLR, CRP, IL‐8, and IL‐6. Figure 2A provides a visual representation of the variables screened by λ within the LASSO regression model. Figure 2B shows the relationship between the natural logarithm of λ and the LASSO regression coefficients. As λ increases, the coefficients of the independent variables are increasingly compressed, causing variables with minimal effects on the outcome to reduce to zero, thereby decreasing the number of variables in the model. In this study, the optimal model was selected at λ = 0.0117, and the independent variables included in the LASSO regression model were operation duration, tumor size, HBP, NLR, and CRP. These findings align with the results of the logistic regression analysis.

3.4
Correlation Analysis Among Different Inflammatory Factors
Spearman's rank correlation analysis revealed a significant positive correlation between HBP and WBC (r = 0.262, p < 0.001), NEUT (r = 0.245, p < 0.001), CRP (r = 0.215, p < 0.001), and IL‐6 (r = 0.130, p < 0.05) (Figure 3). No significant correlation was observed between HBP and NLR (r = 0.042, p > 0.05) or IL‐8 (r = 0.089, p > 0.05). Additionally, WBC showed a significant positive correlation with CRP (r = 0.156, p < 0.01) and IL‐6 (r = 0.137, p < 0.05), but not with IL‐8 (r = 0.04, p > 0.05). NEUT was positively correlated with CRP (r = 0.137, p < 0.05) and IL‐6 (r = 0.152, p < 0.001), but not with IL‐8 (r = 0.036, p > 0.05). NLR demonstrated a significant positive correlation with IL‐6 (r = 0.200, p < 0.001), though no correlation with CRP (r = 0.012, p > 0.05) or IL‐8 (r = 0.078, p > 0.05) was observed. CRP levels were positively correlated with IL‐6 (r = 0.137, p < 0.05) and IL‐8 (r = 0.346, p < 0.001). A significant positive correlation was also found between IL‐8 and IL‐6 (r = 0.134, p < 0.05). Correlation analyses support HBP's role in shared inflammatory pathways. See Figure 3 for details.

3.5
Analysis of the Diagnostic Value of HBP in Pulmonary Infection After Radical Lung Cancer Surgery
We further assess the performance of a panel of inflammatory biomarkers, including HBP, WBC, NEUT, NLR, CRP, IL‐8, and IL‐6, in predicting postoperative pulmonary infection. The optimal cutoff values identified for HBP, WBC, NEUT, NLR, CRP, IL‐6, and IL‐8 in diagnosing postoperative pulmonary infection were 52.36 ng/mL, 13.45 × 109/L, 12.15 × 109/L, 16.77, 24.99 mg/L, 37.02 pg/mL, and 35.40 pg/mL, respectively. The corresponding AUCs were 0.78 (95% CI: 0.69–0.87), 0.71 (95% CI: 0.61–0.81), 0.72 (95% CI: 0.61–0.82), 0.74 (95% CI: 0.65–0.83), 0.72 (95% CI: 0.61–0.83), 0.67 (95% CI: 0.56–0.78), and 0.72 (95% CI: 0.63–0.81), respectively. Combined diagnostic models showed improved predictive accuracy, with AUCs for HBP + WBC, HBP + NLR, HBP + CRP, CRP + NLR, HBP + WBC + CRP, and HBP + NLR + CRP at 0.81 (95% CI: 0.73–0.90), 0.84 (95% CI: 0.76–0.93), 0.85 (95% CI: 0.78–0.92), 0.87 (95% CI: 0.80–0.93), 0.88 (95% CI: 0.82–0.93), and 0.92 (95% CI: 0.89–0.94), respectively. Notably, the AUCs for combined models were consistently higher than those for individual biomarkers. The combination of HBP, NLR, and CRP showed superior diagnostic performance with the highest AUC of 0.92. See Figure 4 and Table 3 for details.

4
Discussion
Postoperative pulmonary infection is a prevalent complication following radical lung resection for lung cancer, contributing to prolonged recovery, extended hospitalization, and increased mortality. A recent meta‐analysis reported an overall incidence of 18.4% for lung complications following lung cancer surgery, with pulmonary infections comprising a significant portion [24]. In our cohort, pulmonary infection incidence was 10.6%, underscoring the critical need for timely prediction and assessment in this population.
HBP, an inflammatory biomarker released by neutrophils, has demonstrated diagnostic value across diverse infectious conditions. During infection, neutrophil‐derived HBP binds to endothelial cell‐adhesion molecules, increasing vascular permeability and facilitating neutrophil recruitment. It also activates monocytes, macrophages, and lymphocytes, contributing to vascular leakage and inflammatory edema [25]. HBP's diagnostic value stems from rapid degranulation kinetics upon pathogen exposure (within minutes) [9, 16], establishing it as a key biomarker for early bacterial infection detection. By contrast, CRP, PCT, and SAA require gene transcription and protein synthesis, resulting in delayed elevation (hours) [16, 17, 18, 19]. Although established inflammatory markers (WBC, CRP, TNF‐α, IL‐6, and SAA) retain utility in early infection detection [26, 27], emerging evidence supports the diagnostic enhancement achieved through combining HBP with conventional biomarkers [28, 29, 30, 31].
Emerging evidences further implicate HBP in pulmonary inflammatory processes. Xue et al. demonstrated that HBP elevation precedes increases in CRP, PCT, and SAA during acute exacerbations of interstitial lung disease (ILD). Their findings suggest HBP's potential utility in assessing disease progression, prognostic stratification, and identification of co‐infections in ILD patients [16]. Additionally, Paulsson et al. identified HBP concentrations in lower airway specimens as a promising diagnostic biomarker for ventilator‐associated pneumonia (VAP) [32]. Capitalizing on rapid HBP quantification (TAT < 2 h) to enable same‐day clinical interventions, our study integrated HBP with complementary inflammatory markers. This combinatorial approach enhances early detection of pulmonary infections following radical lung resection, ultimately optimizing postoperative management and therapeutic decision‐making to improve patient recovery trajectories.
The pathogenic bacteria identified in patients with pulmonary infections following radical lung cancer surgery were predominantly Gram‐negative, with Klebsiella pneumoniae and Escherichia coli as the most frequently isolated species. Notably, these pathogens exhibited high resistance rates to empirical antibiotics [33], underscoring the imperative for vigilant postoperative surveillance and optimized prophylaxis in radical lung cancer surgery. Among patients in our pulmonary infection cohort, 15 exhibited positive cultures, comprising seven cases of Gram‐negative bacterial infections, four of Gram‐positive bacterial infections, and four fungal infections. Subgroup analyses stratified by infection type revealed no statistically significant differences in biomarker levels (including HBP) between groups (See Table S1 for details). This null finding likely reflects limited statistical power due to sample size constraints. Cai et al. demonstrated that HBP and PCT effectively differentiate bacterial and fungal community‐acquired pneumonia (CAP) from viral CAP [29]. Future studies with expanded sample sizes need to investigate HBP's discriminative capacity for specific pulmonary infection pathogens. The NLR, reflecting systemic immune dysregulation, serves as a discriminative tool: elevated NLR typically indicates bacterial/fungal infections, whereas lymphocytic predominance suggests viral etiology [34]. Consistent with this paradigm, the predominance of bacterial/fungal infections in our cohort corresponded to significantly higher NLR in the infection group versus controls.
Logistic and LASSO regression analyses identified HBP, NLR, and CRP as independent predictors of postoperative pulmonary infection. Elevated levels of these inflammatory markers were significantly associated with postoperative pulmonary infection occurrence. Surgical intervention activates the immune response, causing tissue damage and the subsequent release of numerous inflammatory mediators and cytokines, which promote HBP synthesis and release [35]. Infection induces further immune activation and mediator release. Consistently, HBP, NEUT, WBC, CRP, IL‐6, and IL‐8 levels were significantly elevated in the infection group versus controls (p < 0.001), reflecting enhanced inflammatory responses characteristic of infectious pathophysiology. Supporting its clinical relevance, Adam Linder et al. [36] demonstrated that HBP concentration is associated with prognosis, with early HBP elevation in severe cases predicting circulatory failure. In our study, among the 31 infected patients, exploratory subgroup analysis stratified by severity revealed higher inflammatory marker trends in complicated pneumonia cases (n = 4; defined by severe complications including heart failure, sepsis, or ARDS) versus simple cases (n = 27). However, these differences lacked statistical significance (all p values > 0.05; Table S1), likely due to limited statistical power from the small sample of the complicated pneumonia subgroup.
Correlation analysis revealed a significant positive association between HBP, CRP, and IL‐6, suggesting that these markers may be regulated by common mechanisms during infection after lung cancer surgery. Literature suggests that neutrophils release HBP as part of the inflammatory cascade, which then plays a key role in activating and attracting inflammatory cells, leading to the secretion of proinflammatory cytokines like IL‐6 [8, 16]. This process is essential in the coordinated immune response to infection and inflammation. Additionally, cytokines such as IL‐6, IL‐1β, and TNF stimulate the hepatic synthesis and secretion of CRP [37]. Our findings implicate HBP in shared pathogenic mechanisms and provide a mechanistic foundation for future investigations.
ROC analysis indicated that HBP achieved a higher AUC (0.78) than IL‐6 (0.67), IL‐8 (0.72), and WBC (0.71), and performed comparably to CRP (0.72) and NLR (0.74) in diagnosing postoperative pulmonary infection. Given its rapid release kinetics [9, 16], HBP holds theoretical advantages for early infection detection over slower‐responding biomarkers such as CRP and IL‐6. Although the discriminative ability of HBP alone was moderate, its integration with NLR and CRP substantially enhanced diagnostic performance, yielding an AUC of 0.92 (95% CI: 0.89–0.94), which surpassed all individual markers and other biomarker combinations. These findings suggest that while HBP offers valuable independent diagnostic information, its clinical utility is optimized when incorporated into a multi‐marker panel. Such a panel may facilitate earlier and more accurate detection of pulmonary infection, particularly in postoperative settings where inflammatory backgrounds can confound single‐marker interpretation. Further studies are warranted to validate optimal cutoff values for combined biomarkers against microbiological standards and to evaluate their potential in guiding antimicrobial stewardship.
Several limitations should be considered when interpreting our results. First, the relatively small number of infection cases (n = 31) in this retrospective study may limit its statistical power and, further, the generalizability of the findings. Second, the study population was derived from a single tertiary care center, which may introduce selection bias. Future large‐scale, multicenter, prospective studies are needed to validate the diagnostic and clinical utility of HBP, both alone and in combination with other biomarkers, in patients undergoing radical lung cancer surgery.
In conclusion, combined measurement of HBP, NLR, and CRP significantly enhances diagnostic accuracy for pulmonary infection after radical lung cancer surgery. Currently, the clinical application of HBP in infection management is underutilized, and further research is needed to confirm its diagnostic efficacy. Comprehensive biomarker assessment offers a deeper understanding of the causes of pulmonary infections, enabling more accurate and individualized diagnostic and treatment strategies. Further studies should explore additional biomarkers to improve diagnosis and treatment strategies for postoperative pulmonary infections, aiming to reduce complications and enhance patient recovery.

Author Contributions
Conceptualization by Changqiang Chen. Methodology by Yulan Wang and Lulu Guo. Software by Yulan Wang and Bingjie Zeng. Validation by Yulan Wang and Changqiang Chen. Formal analysis by Yulan Wang. Investigation by Yulan Wang, Lulu Guo, Yangyang Xu, Biao Xiang, Ziling Xia, Tianhuan Wang, and Shihou Zheng. Resources by Yulan Wang. Data curation by Yulan Wang and Changqiang Chen. Writing – original draft preparation by Yulan Wang. Writing – review and editing by Changqiang Chen. Visualization by Yulan Wang. Supervision by Changqiang Chen. Project administration by Changqiang Chen. Funding acquisition by Changqiang Chen and Yulan Wang. All authors have read and approved the final version of the manuscript. Changqiang Chen had full access to all of the data in this study and takes complete responsibility for the integrity of the data and the accuracy of the data analysis.

Ethics Statement
This study was conducted in compliance with the Declaration of Helsinki and received approval from the Ethics Review Committee of Shanghai Chest Hospital, Shanghai Jiao Tong University School of Medicine (IRB, IS24124). The ethics committee waived the need for written informed consent from the participants due to the retrospective nature of this study.

Conflicts of Interest
The authors declare no conflicts of interest.

Transparency Statement
The corresponding author, Changqiang Chen, affirms that this manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant, registered) have been explained.

Supporting information

Table S1: Subgroup analyses stratified by severity of pneumonia and infection type.