Abstract
Background
Postoperative Pulmonary Infection (POPI) is a leading cause of morbidity and mortality following esophagectomy. Current static inflammatory markers lack the specificity to distinguish between surgical stress and early infection. This study aims to investigate whether the dynamic perioperative changes in the Systemic Inflammation Response Index (SIRI) can reflect the host’s “inflammation-immune” balance and predict the occurrence of POPI.
Methods
A two-center retrospective cohort study was conducted on 543 patients undergoing radical esophagectomy between January 2020 and November 2025. The SIRI Ratio was defined as postoperative (≤48h) SIRI divided by preoperative SIRI. Univariate and fully adjusted multivariate logistic regression models were performed. Diagnostic value was evaluated using ROC curve analysis, stratified by Neoadjuvant Chemoimmunotherapy (nICT) status.
Results
The overall incidence of POPI was 29.7% (161/543). Patients with POPI exhibited a significantly lower SIRI Ratio (P = 0.002), indicating a blunted acute inflammatory response. Multivariate analysis identified the SIRI Ratio as an independent protective factor against POPI (OR = 0.98, 95% CI: 0.96–0.99, P = 0.037). Notably, a significant interaction was observed between nICT and the SIRI Ratio (P for interaction = 0.004). The SIRI Ratio showed better discrimination in the nICT-naïve subgroup (AUC = 0.776) but its diagnostic value was compromised in the nICT subgroup (AUC = 0.585).
Conclusion
Perioperative changes in SIRI were associated with POPI. The SIRI Ratio showed moderate discrimination overall, with better performance in patients who had not received nICT than in those who had.
Keywords: esophageal cancer, SIRI ratio, postoperative pulmonary infection, neoadjuvant chemoimmunotherapy, Inflammatory response
Graphical Abstract
Introduction
Among the numerous complications following esophagectomy, Postoperative Pulmonary Complications (PPCs), specifically Postoperative Pulmonary Infection (POPI), exhibit the highest incidence and cause the most profound clinical harm. According to the literature, the incidence of POPI fluctuates between 15% and 40%, depending on diagnostic criteria and surgical approaches.1,2
POPI is a key determinant of adverse perioperative outcomes. First, it acts as an independent risk factor for perioperative mortality. Research data indicates that in patients with severe pulmonary infections, in-hospital mortality can soar from less than 1% in non-infected patients to 10%-17%.2 Respiratory failure, sepsis, and subsequent Multiple Organ Dysfunction Syndrome (MODS) induced by infection constitute the primary pathways to death.3–5 Second, accumulating evidence suggests that early postoperative infectious complications exert a long-term “legacy effect,” severely compromising oncological prognosis. POPI has been confirmed as an independent negative predictor of 5-year Overall Survival (OS) and Disease-Free Survival (DFS).3,6,7 The biological basis for this association may lie in the persistent systemic inflammatory response induced by severe infection, which leads to immunosuppression, thereby activating dormant micrometastases or creating a favorable microenvironment for the colonization of Circulating Tumor Cells.7,8 Furthermore, POPI significantly increases the health economic burden. This is reflected not only in prolonged ICU stays and total hospital lengths of stay but also in substantially increased medical costs; previous studies have noted that infection can result in tens of thousands of RMB in additional expenditures.1
Given the severe consequences of POPI, precise identification of high-risk patients preoperatively or in the early postoperative period is crucial. However, current clinical prediction tools have significant limitations. Traditional inflammatory markers such as C-reactive protein (CRP) and White Blood Cell count (WBC), while sensitive, lack specificity. Following highly traumatic procedures like esophagectomy, nearly all patients experience a physiological elevation in CRP and WBC, making it difficult for clinicians to distinguish between a normal surgical stress response and early pathological infection.9 Moreover, existing clinical risk scoring systems (eg, ASA score, Charlson Comorbidity Index) primarily focus on the patient’s static physiological reserve and fail to capture the dynamic changes in immune status during the perioperative period. Therefore, the search for a novel biomarker capable of reflecting the body’s “inflammation-immune” balance in real-time has become a hotspot in current surgical research.
To deeply understand the pathophysiological basis of prediction markers, one must consider the dynamic shifts in immune homeostasis during the perioperative period. Surgery, as a form of controlled severe trauma, triggers a massive release of Damage-Associated Molecular Patterns (DAMPs), immediately activating the innate immune system and initiating a Systemic Inflammatory Response Syndrome (SIRS) driven by neutrophils and monocytes.10 While this inflammatory response is essential for defense, the immense stress of esophageal cancer surgery often disrupts the delicate balance between pro-inflammatory and compensatory anti-inflammatory mechanisms. This disruption can tip the host into a state of relative immunosuppression or “immune paralysis” in the early postoperative phase.11,12 Within this window of immune disequilibrium (typically postoperative days 3–7), the patient’s defense capability against pathogens is compromised, creating a significant vulnerability to pulmonary infections.
Materials and Methods
Study Design and Participants
This retrospective cohort study was conducted at the Department of Thoracic Surgery, Chong Gang General Hospital and Department of Cardiothoracic Surgery, The First Affiliated Hospital of Chongqing Medical University. We initially screened a total of 1043 consecutive patients diagnosed with esophageal cancer between January 2020 and November 2025.
To ensure a homogeneous study cohort and minimize selection bias, strict inclusion and exclusion criteria were applied (Figure 1).
Figure 1.
Workflow for Patient Inclusion.
Inclusion Criteria: Patients were considered eligible if they met the following criteria:
-
(1)
Histological Confirmation: Primary esophageal squamous cell carcinoma (ESCC) or adenocarcinoma (EAC).
-
(2)
Surgical Procedure: Elective radical esophagectomy (McKeown or Ivor Lewis procedure) combined with standard two- or three-field lymph node dissection.
-
(3)
Tumor Staging: Pathological Stage II or III (according to the 8th edition AJCC/UICC TNM staging system). This population was selected to focus on locally advanced disease commonly involving neoadjuvant therapy, thereby enhancing comparability.
Exclusion Criteria: We excluded patients based on the following stepwise criteria:
-
(1)
Ineligibility for Radical Surgery: Patients who underwent palliative surgery, exploratory thoracotomy only, or had distant metastasis (Stage IV) confirmed intraoperatively.
-
(2)
Early-Stage Disease: Patients with Stage I disease were excluded. Rationale: These patients often undergo endoscopic resection or minimally invasive surgery with limited trauma and rarely receive nICT, representing a distinct low-risk population.
-
(3)
Active Preoperative Infection: Clinical evidence of infection or fever >38°C within 14 days prior to surgery.
-
(4)
Incomplete Clinical Data: Defined as the absence of paired blood routine data (specifically neutrophil, lymphocyte, and monocyte counts) within the strict sampling windows (preoperative ≤ 3 days or postoperative ≤ 48 hours), preventing the calculation of the SIRI Ratio.
-
(5)
Early Mortality: Patients who died within 72 hours postoperatively. Rationale: Early death precludes the observation of the primary outcome (POPI) and introduces competing risk bias.
Following this screening process, 500 patients were excluded, and a final cohort of 543 eligible patients was included in the analysis.
No formal a priori sample-size calculation was performed because this retrospective study included all eligible consecutive patients within the prespecified study period. The final cohort comprised 543 patients, including 161 POPI events. Multivariable analyses involving operation duration used 540 complete cases after excluding 3 patients with missing operation-duration data. The final SIRI Ratio multivariable model contained 13 estimated parameters, corresponding to approximately 12 events per parameter.
(6) Prior Radiotherapy: Patients who received neoadjuvant radiotherapy or chemoradiotherapy were explicitly excluded. Rationale: Radiation significantly alters bone marrow function and causes profound, long-lasting lymphopenia, which would severely confound the baseline calculation and dynamic changes of the SIRI.
The study was approved by the Ethics Committee of The First Affiliated Hospital of Chongqing Medical University (Approval No. 2025-559-01) and the Ethics Committee of Chong Gang General Hospital (Approval No. 2025-SY-10), and was conducted in accordance with the ethical principles of the World Medical Association Declaration of Helsinki. The requirement for informed consent was waived by both ethics committees because of the retrospective nature of the study.
Data Collection
Clinical information of the patients was retrospectively reviewed from the electronic medical record system, including age, sex, body mass index (BMI), smoking and drinking history, and comorbidities (hypertension, diabetes, hyperlipidemia, COPD, arrhythmia). Preoperative lung function parameters (FEV1, FVC, FEV1/FVC) were recorded. Surgical details, including operation duration, nICT status (defined as four cycles of camrelizumab combined with nab-paclitaxel and cisplatin), surgical approach (MIE vs Open), and clinical tumor stage were also documented.
The neoadjuvant chemoimmunotherapy regimen consisted of camrelizumab combined with nab-paclitaxel and cisplatin for four 21-day cycles. Surgery was performed 4–6 weeks after the final cycle. The complete dosing schedule is provided in the Supplementary Methods.
A standardized perioperative pathway covered preoperative respiratory preparation, total intravenous anesthesia with one-lung ventilation, multimodal analgesia, ICU/HDU observation, tracheal extubation, early mobilization, chest-tube removal, pulmonary physiotherapy, and antibiotic prophylaxis and escalation. The complete protocol and objective criteria are provided in the Supplementary Methods.
Peripheral blood samples were collected at two time points:
1. Preoperative: Within 3 days prior to surgery.
2. Postoperative: Within 48 hours after surgery.
Routine blood parameters, including neutrophil, monocyte, and lymphocyte counts, were analyzed. The Systemic Inflammation Response Index (SIRI) was calculated using the following formula:
SIRI = absolute neutrophil count (109/L) × Monocyte count (109/L)/Lymphocyte count (109/L)
The SIRI Ratio was defined as the ratio of postoperative SIRI to preoperative SIRI:
SIRI Ratio = Postoperative SIRI/Preoperative SIRI
Additionally, ΔSIRI was calculated for each patient as postoperative SIRI minus preoperative SIRI.
Definition of POPI
The primary outcome of this study was the occurrence of POPI during the hospitalization period. POPI was diagnosed according to the criteria established by the Chinese guidelines for hospital-acquired pneumonia and the US Centers for Disease Control and Prevention (CDC) criteria. The diagnosis typically required the presence of at least two of the following: (1) Radiological evidence: New or progressive infiltrates, consolidation, or ground-glass opacities on chest X‑ray or CT scan, excluding non‑infectious etiologies such as atelectasis or pulmonary edema. (2) Clinical signs: Fever (body temperature >38°C) with no other apparent source, along with new or worsening cough, purulent sputum, dyspnea, or auscultatory findings of moist rales. (3) Laboratory findings: Leukocytosis (white blood cell count >12×109/L) or leukopenia (<4×109/L), or a significant rise in inflammatory markers (eg, C‑reactive protein >50 mg/L or procalcitonin >0.5 ng/mL). (4) Microbiological support (optional but reinforcing): Isolation of a likely pathogen from sputum, bronchoalveolar lavage fluid, or blood culture, or detection of pathogen nucleic acids via molecular assays. Cases with preoperative active pulmonary infection, transient postoperative atelectasis or inflammation resolving within 72 h, or non‑infectious pulmonary complications (eg, acute respiratory distress syndrome) were excluded. The diagnostic time window spanned from the day of surgery until hospital discharge.
Statistical Analysis
Statistical analyses were performed using R software version 4.3.0. Continuous variables were checked for normality using the Shapiro–Wilk test. Normally distributed data were expressed as mean ± standard deviation (SD) and compared using the Student’s t-test. Non-normally distributed data were presented as median [interquartile range, IQR] and compared using the Mann–Whitney U-test. Categorical variables were expressed as frequencies (percentages) and compared using the Chi-square test or Fisher’s exact test. Analyses involving operation duration used complete cases (n = 540); three patients with missing operation-duration data were excluded.
Univariate logistic regression was performed to evaluate the association of the SIRI metrics with POPI. Multivariable models evaluated prespecified SIRI exposures using clinically selected covariates. Because Preoperative SIRI, Postoperative SIRI, ΔSIRI, and the SIRI Ratio are mathematically related, they were not entered simultaneously in the primary adjusted models. The primary models evaluated Preoperative SIRI and the SIRI Ratio separately; ΔSIRI was evaluated in supplementary adjusted models.
Model 1 was adjusted for age, sex, smoking, and alcohol history.
Model 2 was fully adjusted for age, sex, smoking, alcohol, hypertension, diabetes, FEV1, operation duration, nICT, surgical approach, clinical tumor stage, and study center.
Results were presented as odds ratios (ORs) with 95% confidence intervals (CIs). Unless otherwise stated, ORs for continuous SIRI metrics represent the change per 1-unit increase. Mann–Whitney U-tests compare rank distributions between groups, whereas logistic regression estimates a specified association with the log odds of infection.
Receiver Operating Characteristic (ROC) curve analysis was conducted to evaluate the predictive value of SIRI metrics. The Area Under the Curve (AUC) was calculated, and the optimal cutoff values were determined based on the maximum Youden index (Sensitivity + Specificity - 1).
Subgroup analyses were performed based on relevant clinical characteristics, and interactions were tested to assess potential effect modifications.
To visualize the dose-response relationship between the SIRI Ratio and the risk of POPI, Restricted Cubic Spline (RCS) analysis was performed with 4 knots at the 10th, 35th, 65th, and 90th percentiles. Non-linearity was tested by comparing the model with the linear model using the likelihood ratio test. A two-sided P-value < 0.05 was considered statistically significant for all analyses.
Calibration, bootstrap validation, and clinical utility analyses were additionally performed for the SIRI Ratio. Calibration was assessed using a calibration plot, calibration intercept and slope, and the Brier score. Internal validation used 1,000 bootstrap resamples with a fixed random seed. Decision-curve analysis estimated net benefit across threshold probabilities of 0.05–0.50 and compared the SIRI Ratio model with WBC, treat-all, and treat-none strategies. The SIRI Ratio and WBC ROC curves were compared using the paired DeLong test. To benchmark the SIRI Ratio against established inflammatory biomarkers, paired ROC analyses compared it with the perioperative NLR Ratio in the full cohort and with postoperative procalcitonin in complete cases with available measurements. AUC 95% confidence intervals and paired comparisons were calculated using the DeLong method. Detailed methods and results are provided in the Supplementary Statistical Methods, Supplementary Tables S1, S2, and Supplementary Figures S1–S3. To clarify the scale dependence of postoperative SIRI, univariable logistic regression was repeated on the raw and log2-transformed scales. Separate supplementary multivariable models evaluated ΔSIRI without simultaneously including preoperative SIRI, postoperative SIRI, or the SIRI Ratio. Model 1 adjusted for age, sex, smoking, and alcohol; Model 2 additionally adjusted for study center, hypertension, diabetes, FEV1, operation duration, surgical approach, clinical stage, and nICT.
Result
Baseline Characteristics of the Study Population
A total of 543 patients who underwent radical esophagectomy were included in this study. Postoperative pulmonary infection occurred in 161 patients, corresponding to an incidence rate of 29.7%. Given the multicenter design of this study, we first compared the baseline demographic and clinical characteristics between patients from the two participating institutions (Supplementary Table S3). Notably, no statistically significant differences were observed between The First Affiliated Hospital of Chongqing Medical University (n = 432) and Chong Gang General Hospital (n = 111) regarding age, sex, comorbidities, pulmonary function (FEV1), neoadjuvant chemoimmunotherapy status, surgical approach, or the primary outcome of POPI incidence (all P > 0.05). The absence of detected between-center differences supported pooling the two cohorts for the present exploratory analyses. The baseline demographic and clinical characteristics of the study cohort are summarized in Table 1.
Table 1.
Baseline Demographic and Clinical Characteristics of Patients
| Level | Overall | Infection | No Infection | p | |
|---|---|---|---|---|---|
| 543 | 161 | 382 | |||
| Age (mean (SD)) | 64.04 (7.67) | 64.20 (7.88) | 63.98 (7.59) | 0.752 | |
| Sex (%) | Female | 86 (15.8) | 19 (11.8) | 67 (17.5) | 0.123 |
| Male | 457 (84.2) | 142 (88.2) | 315 (82.5) | ||
| Study Center (%) | CQMU | 432 (79.6) | 124 (77) | 308 (80.6) | 0.403 |
| CGG | 111 (20.4) | 37 (23.0) | 74 (19.4) | ||
| BMI (mean (SD)) | 22.64 (2.94) | 22.53 (2.77) | 22.69 (3.01) | 0.568 | |
| Smoking (%) | No | 208 (38.3) | 54 (33.5) | 154 (40.3) | 0.166 |
| Yes | 335 (61.7) | 107 (66.5) | 228 (59.7) | ||
| Alcohol (%) | No | 220 (40.5) | 58 (36.0) | 162 (42.4) | 0.198 |
| Yes | 323 (59.5) | 103 (64.0) | 220 (57.6) | ||
| Hypertension (%) | No | 424 (78.1) | 128 (79.5) | 296 (77.5) | 0.685 |
| Yes | 119 (21.9) | 33 (20.5) | 86 (22.5) | ||
| Diabetes (%) | No | 496 (91.3) | 143 (88.8) | 353 (92.4) | 0.234 |
| Yes | 47 (8.7) | 18 (11.2) | 29 (7.6) | ||
| Hyperlipidemia (%) | No | 405 (74.6) | 128 (79.5) | 277 (72.5) | 0.109 |
| Yes | 138 (25.4) | 33 (20.5) | 105 (27.5) | ||
| COPD (%) | No | 444 (81.8) | 124 (77.0) | 320 (83.8) | 0.082 |
| Yes | 99 (18.2) | 37 (23.0) | 62 (16.2) | ||
| Arrhythmia (%) | No | 364 (67.0) | 106 (65.8) | 258 (67.5) | 0.776 |
| Yes | 179 (33.0) | 55 (34.2) | 124 (32.5) | ||
| Clinical Stage (%) | Stage II | 229 (42.2) | 73 (45.3) | 156 (40.8) | 0.381 |
| Stage III | 314 (57.8) | 88 (54.7) | 226 (59.2) | ||
| FVC (mean (SD)) | 3.52 (0.76) | 3.48 (0.71) | 3.53 (0.79) | 0.516 | |
| FEV1 (mean (SD)) | 2.51 (0.64) | 2.49 (0.61) | 2.52 (0.65) | 0.579 | |
| FEV1_FVC (mean (SD)) | 71.33 (8.94) | 71.34 (9.11) | 71.33 (8.88) | 0.992 | |
| nICT status (%) | nICT | 331 (61.0) | 117 (72.7) | 214 (56.0) | <0.001 |
| Non-nICT | 212 (39.0) | 44 (27.3) | 168 (44.0) | ||
| Op_Approach (%) | MIE | 326 (60.0) | 69 (42.9) | 257 (67.3) | <0.001 |
| Open | 217 (40.0) | 92 (57.1) | 125 (32.7) | ||
| Op_Duration (median [IQR]; n=540) | 315.00 [250.00, 388.50] | 310.00 [242.50, 402.50] | 317.00 [255.00, 375.00] | 0.452 | |
| Preoperative SIRI (median [IQR]) | 0.97 [0.65, 1.34] | 1.00 [0.73, 1.56] | 0.95 [0.61, 1.24] | 0.649 | |
| Postoperative SIRI (median [IQR]) | 10.52 [6.45, 15.80] | 6.51 [4.12, 9.31] | 12.54 [8.24, 17.74] | <0.001 | |
| ΔSIRI (median [IQR]) | 9.38 [5.38, 14.86] | 5.25 [2.91, 8.16] | 11.54 [7.34, 16.85] | <0.001 | |
| Ratio SIRI (median [IQR]) | 10.66 [6.42, 16.71] | 10.07 [5.76, 15.01] | 14.09 [8.74, 20.08] | 0.002 |
Abbreviations: CQMU, The First Affiliated Hospital of Chongqing Medical University; CGG, Chong Gang General Hospital; SD, Standard Deviation; BMI, Body Mass Index; FEV1, Forced Expiratory Volume in 1 second; MIE, Minimally Invasive Esophagectomy; Op-, Operation-.
There were no statistically significant differences between the infection and non-infection groups regarding age, sex, body mass index (BMI), smoking and drinking history, or comorbidities such as hypertension, diabetes, and COPD (all P > 0.05). Regarding surgical variables, operation duration was comparable between the two groups (P = 0.452). However, the surgical approach differed significantly; the proportion of patients undergoing open thoracotomy was higher in the infection group than in the non-infection group (57.1% vs 32.7%, P < 0.001). The proportion receiving nICT was also higher in the infection group (72.7% vs 56.0%, P < 0.001). Preoperative SIRI was comparable between the infection and non-infection groups (1.00 [0.73, 1.56] vs 0.95 [0.61, 1.24], P = 0.649). In contrast, postoperative SIRI was lower in the infection group (6.51 [4.12, 9.31] vs 12.54 [8.24, 17.74], P < 0.001), and the perioperative increase was smaller (ΔSIRI: 5.25 [2.91, 8.16] vs 11.54 [7.34, 16.85], P < 0.001). The Post/Pre SIRI Ratio was likewise lower in the infection group (10.07 [5.76, 15.01] vs 14.09 [8.74, 20.08], P = 0.002), consistent with a blunted postoperative inflammatory response.
Given the potential impact of neoadjuvant therapy on systemic inflammation, we further compared SIRI metrics between the nICT and non-nICT groups (Figure 2). Patients in the nICT group exhibited significantly lower Pre-SIRI levels (P < 0.001), significantly lower Post-SIRI levels (P < 0.001), and a lower SIRI Ratio (P = 0.0019) than patients in the non-nICT group.
Figure 2.
Comparison of perioperative SIRI metrics between the nICT and non-nICT groups. (A) Preoperative SIRI. (B) Postoperative SIRI. (C) Post/Pre SIRI Ratio. P values were calculated using the Mann–Whitney U-test.
Abbreviation: SIRI, Systemic Inflammation Response Index.
Univariate Logistic Regression Analysis of SIRI Metrics
Univariate logistic regression was used to evaluate Preoperative SIRI, Postoperative SIRI, ΔSIRI, and the SIRI Ratio in relation to postoperative pulmonary infection (Table 2). Preoperative SIRI was associated with higher odds of infection (OR = 1.294, 95% CI: 1.017–1.647, P = 0.036). In contrast, both ΔSIRI (OR = 0.829, 95% CI: 0.702–0.931, P = 0.009) and the Post/Pre SIRI Ratio (OR = 0.780, 95% CI: 0.661–0.999, P = 0.018) were associated with lower odds of infection.
Table 2.
Univariate Logistic Regression Analysis of SIRI Metrics for Predicting Pulmonary Infection
| Variable | OR (95% CI) | P_Value |
|---|---|---|
| Pre_SIRI (per 1-unit increase) | 1.294 (1.017, 1.647) | 0.036 |
| Post_SIRI (per 1-unit increase) | 0.842 (0.706, 1.068) | 0.103 |
| ΔSIRI (per 1-unit increase) | 0.829 (0.702, 0.931) | 0.009 |
| SIRI Ratio (per 1-unit increase) | 0.780 (0.661, 0.999) | 0.018 |
Notes: ORs for continuous SIRI metrics represent the change per 1-unit increase.
Abbreviations: CI, confidence interval; OR, odds ratio; SIRI, Systemic Inflammation Response Index.
Postoperative SIRI did not reach statistical significance when modeled linearly on the raw scale (OR per 1-unit increase = 0.842, 95% CI: 0.706–1.068, P = 0.103). Because postoperative SIRI was markedly right-skewed, a log2-transformed sensitivity analysis was performed; each two-fold increase was associated with lower odds of infection (OR = 0.314, 95% CI: 0.245–0.403, P < 0.001; Supplementary Table S4). Although ΔSIRI differed between groups and was significant in univariate analysis, it was not independently associated with infection after adjustment in Model 1 (adjusted OR = 0.981, 95% CI: 0.962–1.000, P = 0.052) or Model 2 (adjusted OR = 0.973, 95% CI: 0.933–1.014, P = 0.197; Supplementary Table S5). Subsequent diagnostic and subgroup analyses therefore remained focused on the SIRI Ratio as the principal baseline-normalized dynamic metric.
Multivariate Logistic Regression Analysis of Pre_SIRI and SIRI Ratio for Postoperative Pulmonary Infection
Multivariate logistic regression analysis was performed to evaluate the associations of Pre_SIRI and SIRI Ratio with POPI. Two models were constructed: Model 1 adjusted for age, sex, smoking, and alcohol; Model 2 further adjusted for hypertension, diabetes, operation duration, FEV1, surgical approach, clinical tumor stage, study center, and nICT group.
For Pre_SIRI, no statistically significant independent association with POPI was observed in either model. In Model 1, the odds ratio (OR) was 1.243 (95% CI: 0.971–1.591, P = 0.084). After fully adjusting for confounding factors in Model 2, the association remained non-significant (OR = 1.131, 95% CI: 0.876–1.459, P = 0.344). In Model 2, nICT was associated with higher odds of POPI (OR = 2.312, P < 0.001), whereas higher FEV1 was protective (OR = 0.650, P = 0.018). Open surgical approach was also associated with higher odds of POPI (OR = 1.462, P = 0.002), while operation duration showed borderline evidence of association (OR = 1.002, P = 0.050) (Table 3).
Table 3.
Multivariate Logistic Regression Analysis of the Association Between Pre_SIRI and Postoperative Pulmonary Infection
| Variable | OR (95% CI) | P_Value |
|---|---|---|
| Model 1 | ||
| Pre_SIRI | 1.243 (0.971, 1.591) | 0.084 |
| Age | 1.004 (0.980, 1.029) | 0.748 |
| Sex | 1.294 (0.665, 2.518) | 0.448 |
| Smoking | 1.096 (0.649, 1.850) | 0.732 |
| Alcohol | 1.086 (0.660, 1.787) | 0.745 |
| Model 2 (n=540) | ||
| Pre_SIRI | 1.131 (0.876, 1.459) | 0.344 |
| Age | 1.000 (0.975, 1.027) | 1.000 |
| Sex | 1.494 (0.741, 3.012) | 0.262 |
| Study Center | 0.982 (0.612, 1.575) | 0.940 |
| Smoking | 1.061 (0.618, 1.821) | 0.830 |
| Alcohol | 0.959 (0.579, 1.588) | 0.871 |
| Hypertension | 0.806 (0.496, 1.310) | 0.384 |
| Diabetes | 1.735 (0.895, 3.366) | 0.103 |
| FEV1 | 0.650 (0.450, 0.920) | 0.018 |
| Op_Duration | 1.002 (1.000, 1.004) | 0.050 |
| Op_Approach | 1.462 (1.187, 1.904) | 0.002 |
| Clinical Stage | 1.354 (0.733, 2.547) | 0.340 |
| nICT (vs non-nICT) | 2.312 (1.496, 3.573) | <0.001 |
Notes: ORs for continuous Pre_SIRI represent the change per 1-unit increase. Model 1 adjusted for age, sex, smoking, and alcohol consumption. Model 2 was fully adjusted for multiple covariates (n = 540).
Abbreviations: CI, confidence interval; OR, odds ratio; Pre_SIRI, preoperative Systemic Inflammation Response Index; nICT, neoadjuvant chemoimmunotherapy.
In contrast, the SIRI Ratio demonstrated a significant association with POPI. In Model 1, a higher SIRI Ratio was significantly associated with a decreased risk of postoperative pulmonary infection (OR = 0.822, 95% CI: 0.721–0.936, P = 0.003). This association remained significant in Model 2 after adjustment for all covariates (OR = 0.98, 95% CI: 0.96–0.99, P = 0.037), suggesting that the SIRI Ratio serves as an independent predictor. In the fully adjusted model, nICT (OR = 1.612, P < 0.001) and open surgical approach (OR = 1.351, P = 0.018) were associated with higher odds of POPI, whereas higher FEV1 was protective (OR = 0.712, P = 0.033) (Table 4).
Table 4.
Multivariate Logistic Regression Analysis of the Association Between SIRI Ratio and Postoperative Pulmonary Infection
| Variable | OR (95% CI) | P_Value |
|---|---|---|
| Model 1 | ||
| Ratio | 0.822 (0.721, 0.936) | 0.003 |
| Age | 1.004 (0.980, 1.029) | 0.748 |
| Sex | 1.326 (0.683, 2.573) | 0.404 |
| Smoking | 1.094 (0.648, 1.845) | 0.736 |
| Alcohol | 1.101 (0.669, 1.810) | 0.705 |
| Model 2 (n=540) | ||
| Ratio | 0.98 (0.96, 0.99) | 0.037 |
| Age | 1.000 (0.975, 1.026) | 1.000 |
| Sex | 1.270 (0.642, 2.510) | 0.492 |
| Study Center | 1.035 (0.658, 1.627) | 0.882 |
| Smoking | 1.205 (0.707, 2.054) | 0.493 |
| Alcohol | 1.023 (0.620, 1.690) | 0.929 |
| Hypertension | 0.759 (0.470, 1.224) | 0.259 |
| Diabetes | 1.677 (0.875, 3.214) | 0.119 |
| FEV1 | 0.712 (0.521, 0.973) | 0.033 |
| Op_Duration | 1.001 (0.999, 1.003) | 0.327 |
| Op_Approach | 1.351 (1.092, 1.799) | 0.018 |
| Clinical Stage | 1.407 (0.507, 2.379) | 0.387 |
| nICT (vs non-nICT) | 1.612 (1.406, 1.913) | <0.001 |
Notes: ORs for continuous SIRI Ratio represent the change per 1-unit increase. Model 1 adjusted for age, sex, smoking, and alcohol consumption. Model 2 was fully adjusted for multiple covariates (n = 540).
Abbreviations: CI, confidence interval; OR, odds ratio; SIRI, Systemic Inflammation Response Index; nICT, neoadjuvant chemoimmunotherapy.
Furthermore, when “Study Center” was incorporated into the fully adjusted multivariable models (Model 2), it did not emerge as a significant independent predictor for POPI (P > 0.05).
Diagnostic Value and Optimal Cutoff of SIRI Ratio in Overall and Subgroup Populations
To clarify the diagnostic capability of the SIRI Ratio for POPI and to determine the optimal cutoff values, receiver operating characteristic (ROC) curve analysis was performed, and the Youden index was calculated for the total population and subgroups stratified by nICT (Table 5). In the total population (Figure 3A and D), the SIRI Ratio demonstrated moderate predictive ability with an area under the curve (AUC) of 0.696 (95% CI: 0.647–0.744). The optimal cutoff value was identified as 14.17, corresponding to a maximum Youden index of 0.322. Subgroup analysis revealed that the predictive performance of the SIRI Ratio was notably influenced by nICT status. In the non-nICT group (Figure 3B and E), the SIRI Ratio showed higher diagnostic performance, with the AUC increasing to 0.776 (95% CI: 0.696–0.855). The optimal cutoff value for this group was 17.84, with a maximum Youden index of 0.491, indicating higher sensitivity and specificity in this specific cohort. In contrast, the diagnostic value of the SIRI Ratio was limited in the nICT group (Figure 3C and F), showing an AUC of only 0.585 (95% CI: 0.520–0.651). The optimal cutoff value was 11.62, with a Youden index of 0.182. These findings suggest that the predictive value of the SIRI Ratio for POPI is most pronounced in the non-nICT subgroup. In contrast, the predictive capability is compromised in the nICT population. These cutoff values are exploratory subgroup estimates and should not be interpreted as externally validated clinical decision thresholds.
Table 5.
Diagnostic Performance of the SIRI Ratio for Predicting Postoperative Pulmonary Infection in Overall and nICT-Stratified Cohorts
| Parameter | Overall | Non-nICT | nICT |
|---|---|---|---|
| AUC | 0.696 | 0.776 | 0.585 |
| 95% CI Lower | 0.647 | 0.696 | 0.52 |
| 95% CI Upper | 0.744 | 0.855 | 0.651 |
| Optimal Cutoff | 14.171 | 17.836 | 11.621 |
| Sensitivity | 0.665 | 0.818 | 0.556 |
| Specificity | 0.657 | 0.673 | 0.626 |
| Youden Index | 0.322 | 0.491 | 0.182 |
| Sample Size | 543 | 212 | 331 |
| Cases | 161 | 44 | 117 |
| Controls | 382 | 168 | 214 |
Notes: Receiver operating characteristic curve analysis was performed stratified by neoadjuvant chemoimmunotherapy status. The optimal cutoff value was identified according to the maximum Youden index.
Abbreviations: AUC, area under the curve; CI, confidence interval; SIRI, Systemic Inflammation Response Index; nICT, neoadjuvant chemoimmunotherapy.
Figure 3.
ROC curve analysis and Youden index plots of the SIRI Ratio for predicting postoperative pulmonary infection. (A and D) Total population. (B and E) non-nICT group. (C and F) nICT group. Red dots indicate the optimal cutoffs selected by the maximum Youden index.
Abbreviations: AUC, area under the curve; CI, confidence interval.
Additional validation analyses showed limited optimism, with an optimism-corrected AUC of 0.695 and an optimism-corrected Brier score of 0.192. Decision-curve analysis showed greater net benefit than the treat-all, treat-none, and preoperative WBC strategies across threshold probabilities of 0.17–0.50. The SIRI Ratio also showed better discrimination than preoperative WBC in a paired ROC comparison (P = 0.0069). Detailed results are presented in Supplementary Table S1, Supplementary Figures S1 and S2.
Direct comparisons with established inflammatory biomarkers showed that, in the full cohort (n = 543), the SIRI Ratio had an AUC of 0.696 (95% CI, 0.647–0.744) compared with 0.676 (95% CI, 0.624–0.728) for the NLR Ratio; the paired difference was not statistically significant (DeLong P = 0.229). Among patients with available procalcitonin measurements (n = 478), the AUCs were 0.689 (95% CI, 0.633–0.744) for the SIRI Ratio and 0.625 (95% CI, 0.575–0.676) for procalcitonin, with no statistically significant difference (DeLong P = 0.099). Thus, the SIRI Ratio showed numerically higher discrimination, but statistical superiority over these comparators was not demonstrated. Detailed results are presented in Supplementary Table S2 and Supplementary Figure S3.
Subgroup Analysis
To explore whether the association between the SIRI Ratio and POPI was consistent across different clinical characteristics, we performed a stratified subgroup analysis (Figure 4).
Figure 4.
Forest plot of subgroup analyses for the association between the SIRI Ratio and postoperative pulmonary infection. Points show ORs and horizontal lines show 95% CIs; the vertical dashed line indicates OR = 1. Red markers denote subgroup associations with P < 0.05 and blue markers denote P ≥ 0.05. The operation-duration subgroup analysis included 540 patients; 3 patients with missing operation-duration data were excluded from this subgroup analysis.
The forest plot showed that, overall, the SIRI Ratio was negatively associated with the risk of POPI (OR=0.98, 95% CI: 0.96–0.99, P=0.037). However, tests for interaction identified clear evidence of effect modification by nICT (P for interaction = 0.004) and borderline evidence for operation duration (P for interaction = 0.047).
Specifically, an inverse association between the SIRI Ratio and POPI was observed in the non-nICT subgroup (OR=0.93, 95% CI: 0.89–0.98, P=0.003). However, no significant association was found in the nICT subgroup (OR=1.01, 95% CI: 0.98–1.03, P=0.672), suggesting that the marker loses its predictive capability in this population.
Furthermore, operation duration showed borderline evidence of effect modification. In patients with an operation duration ≥ 315 min, a higher SIRI Ratio was significantly associated with a reduced risk of infection (OR=0.96, 95% CI: 0.93–0.99, P=0.009). In contrast, no significant association was observed in patients with a shorter operation duration (< 315 min, P=0.734). The operation-duration subgroup analysis included 540 patients; 3 patients with missing operation-duration data were excluded from this subgroup analysis.
For other stratifying factors, including sex, smoking, alcohol consumption, hypertension, diabetes, and COPD, the interaction tests were not statistically significant (all P for interaction > 0.05). This suggests that the association between the SIRI Ratio and POPI remains relatively stable across these subgroups, despite variations in statistical significance within individual strata.
Notably, regarding surgical characteristics, no significant interaction was observed for the surgical approach (MIE vs Open, P for interaction = 0.822) or clinical stage (P = 0.284). No evidence of effect modification by surgical approach or clinical stage was detected.
To evaluate the dose-response relationship between the SIRI Ratio and the risk of POPI, a Restricted Cubic Spline (RCS) model was employed for visual analysis (Figure 5).
Figure 5.
Dose-response relationship between the SIRI Ratio and postoperative pulmonary infection based on restricted cubic spline analysis. The model was adjusted for age, sex, smoking, alcohol, hypertension, diabetes, operation duration, FEV1, nICT, surgical approach, and clinical tumor stage. The solid red line represents estimated risk and the shaded area represents the 95% CI. P for non-linearity = 0.925.
The RCS analysis indicated no significant non-linear deviation (P for non-linear = 0.925). The fitted curve demonstrated a monotonic linear decreasing trend in the probability of POPI as the SIRI Ratio increased (P for overall association = 0.017).
In conclusion, these findings support an approximately linear inverse association between the SIRI Ratio and the risk of POPI, justifying the rationale for treating the SIRI Ratio as a continuous linear variable in the subsequent regression analyses.
Discussion
This study examined the association between perioperative SIRI dynamics and POPI after esophagectomy. In the reported analyses, a higher SIRI Ratio was associated with lower POPI risk, and its discrimination differed by nICT status. Performance was stronger in the non-nICT subgroup (AUC = 0.776) than in the nICT subgroup (AUC = 0.585). The latter value indicates limited discrimination and supports a context-dependent, exploratory interpretation.
Composite indices derived from routine blood cell counts, including SIRI and the systemic immune-inflammation index, have been investigated across metabolic, hepatic, respiratory, and emergency-care settings, supporting their role as accessible but context-dependent markers of systemic inflammation and prognosis.13–16 More broadly, systemic immune competence depends on coordinated cytokine signaling and innate-adaptive crosstalk; dysregulation involving interferon pathways, tumor-associated immunity, and immune aging can coexist with chronic inflammation and impaired host defense.17–19 Experimental studies of traumatic and fibrotic tissue injury further show that macrophage activation and polarization can shape post-injury inflammation and tissue remodeling.20,21
SIRI integrates neutrophil, lymphocyte, and monocyte counts and therefore represents a different inflammatory construct from single-cell measures or NLR. In the available data, the SIRI Ratio showed better discrimination than preoperative WBC (paired DeLong P = 0.0069). Direct paired comparisons showed numerically higher AUCs for the SIRI Ratio than for the NLR Ratio (0.696 vs 0.676) and procalcitonin (0.689 vs 0.625), but neither difference was statistically significant (P = 0.229 and P = 0.099, respectively). These findings support the SIRI Ratio as a potentially complementary inflammatory marker rather than establishing superiority over existing biomarkers. A principal finding of our study is the inverse association between the SIRI Ratio and infection risk (OR = 0.98). Conventionally, hyperinflammation is considered detrimental.22–24 However, in the immediate perioperative period following major trauma like esophagectomy, a robust pro-inflammatory response is physiologically necessary to clear pathogens and initiate healing.22,25,26
Our data suggests that a high SIRI Ratio (representing a sharp increase from baseline) reflects a “healthy” acute stress response, characterized by successful marrow mobilization of neutrophils and monocytes. Conversely, a low SIRI Ratio in the infection group likely signifies a “blunted response”.27,28 In this state, the host immune system fails to mount a sufficient “alarm” reaction to the surgical hit, allowing commensal or nosocomial bacteria to colonize the respiratory tract unchecked. This concept aligns with studies in sepsis and trauma, where the inability to elevate leukocyte counts in response to insults is often a harbinger of poor prognosis.29–31
The interaction between nICT and the SIRI Ratio represents an important exploratory finding of our study. We observed that nICT significantly lowered both the baseline Pre-SIRI and the postoperative SIRI Ratio. In our cohort, neoadjuvant therapy consisted of four cycles of camrelizumab combined with nab-paclitaxel and cisplatin, with radiotherapy explicitly excluded to avoid radiation-induced severe lymphopenia. This aligns with existing literature suggesting that cytotoxic agents, including taxanes and platinum compounds, cause varying degrees of myelosuppression, effectively “resetting” the immunological baseline.32–34 However, contrary to the assumption that this “reset” might enhance sensitivity, our results showed that the SIRI Ratio had higher discrimination in the non-nICT subgroup (AUC 0.776) compared to the nICT subgroup (AUC 0.585). We interpret surgery as a “stress test” for the bone marrow reserve. In non-nICT patients with intact marrow function, the stress response is proportional to the insult; thus, a failure to mobilize (Low Ratio) accurately signals a compromised host defense, which may contribute to the stronger discrimination observed in this subgroup. Conversely, in the nICT subgroup, we propose a “Chemoimmunotherapy-Induced Blunted Response Hypothesis.” On one hand, cytotoxic agents (such as taxanes and platinum) cause varying degrees of myelosuppression, restricting the rapid mobilization of neutrophils and monocytes from the bone marrow.35 On the other hand, the addition of immune checkpoint inhibitors (ICIs) profoundly reshapes the systemic immune landscape.36 While ICIs revitalize exhausted T cells, they also induce complex recalibrations in the myeloid compartment. Evidences suggests that systemic ICI therapy can alter the functional plasticity of monocytes37 and potentially induce a state of immune tolerance or hyper-activation baseline, which may exhaust the patient’s functional reserve to mount a subsequent, proportional acute stress response to surgical trauma.38 Consequently, the combination of marrow suppression and immunotherapy-driven immune recalibration heavily restricts the magnitude of postoperative SIRI elevation. This hypothesized mechanical and functional restriction may help explain why the SIRI Ratio loses its discriminative alarm function in the nICT subset of patients. Although our retrospective design precluded the direct measurement of functional molecular markers via flow cytometry, our reliance on SIRI—a composite index heavily weighted by monocyte and lymphocyte counts—aligns with well-established immunological paradigms. Extensive evidence demonstrates that major surgery and severe trauma trigger a transient phase of profound immunosuppression, or “immunoparalysis”, the hallmark of which is the drastic downregulation of human leukocyte antigen-DR (HLA-DR) on circulating monocytes.39,40 Monocytes act as the critical bridge between innate and adaptive immunity; their functional deactivation and failure to recover HLA-DR expression postoperatively are strongly predictive of an inability to clear pathogens, leading to secondary nosocomial infections.41,42 Crucially, translational studies of postoperative immune dysfunction provide indirect biological support for a possible link between altered leukocyte dynamics and reduced monocyte HLA-DR expression; this mechanism was not directly measured in our cohort. Accordingly, the blunted postoperative SIRI trajectory observed in the nICT cohort may reflect treatment-related immune dysregulation, although this interpretation remains hypothetical. Monocyte HLA-DR has also been evaluated as a prognostic component of a systemic inflammation response-based model in advanced non-small cell lung cancer.43
Furthermore, Our analysis elucidates the distinct yet complementary biological significance of static versus dynamic SIRI metrics. Elevated Pre-SIRI (Static) likely serves as an indicator of the patient’s “baseline inflammatory load,” often correlating with tumor burden or poor nutritional status. Consistent with our univariate analysis, a high static baseline identifies a physiologically vulnerable “starting point.” In contrast, the SIRI Ratio (Dynamic) captures a different dimension: the host’s “functional reserve” to mount an acute defense against surgical trauma. A high Ratio signifies a robust, physiological mobilization of the innate immune system, which is essential for clearing perioperative pathogens.
Clinical Implications
The SIRI Ratio is inexpensive and readily derived from routine blood counts. By capturing perioperative changes in inflammatory status, it reflects a different dimension of the host response from static measurements. Decision-curve analysis suggested potential clinical net benefit relative to treat-all and treat-none strategies across threshold probabilities of 0.17–0.50. The nICT-stratified cutoffs may help generate hypotheses for future validation, but the limited discrimination in the nICT subgroup (AUC = 0.585) precludes their use as treatment thresholds. The operation-duration interaction was borderline (P for interaction = 0.047); therefore, we did not derive duration-specific cutoffs. We also did not construct a patient-level risk calculator because doing so in the same retrospective cohort without independent external validation would risk overfitting. The SIRI Ratio should be considered an exploratory adjunct to clinical assessment, imaging, and established laboratory markers rather than a standalone decision tool.
Several limitations should be acknowledged. First, the retrospective two-center design remains vulnerable to selection bias, residual confounding, and temporal or center-specific practice variation. Second, the study did not include an independent external cohort; similarities between the two participating centers do not replace external validation. Third, the calibration, bootstrap validation, and decision-curve analyses were conducted within the present cohort and require confirmation in an independent external cohort. Fourth, direct comparisons with the NLR Ratio and procalcitonin did not demonstrate statistically significant superiority of the SIRI Ratio; the procalcitonin comparison was restricted to 478 complete cases, and CRP was unavailable for analysis. Fifth, the detailed nICT regimen and standardized perioperative protocol are reported in the Supplementary Methods; nevertheless, unmeasured variation in protocol adherence between centers cannot be excluded. Finally, cytokines and functional immune markers were not measured, so the proposed immune-blunting mechanism remains hypothetical. Prospective, independently validated multicenter studies are required before clinical implementation. In addition, the subgroup-specific cutoffs and interaction findings remain exploratory and require prospective validation before being incorporated into individualized risk prediction. Because postoperative SIRI was measured within 48 hours and POPI was ascertained from surgery through discharge, temporal overlap between biomarker sampling and the earliest infection manifestations cannot be completely excluded.
Conclusion
In conclusion, perioperative changes in SIRI were associated with POPI in this retrospective cohort. The SIRI Ratio showed moderate discriminatory ability in the overall cohort, with stronger performance in patients without nICT but limited discrimination in those who received nICT.
Funding Statement
No funding was received.
Data Sharing Statement
The data used to support the findings of this study are available from the corresponding authors upon request.
Ethical Approval
This retrospective observational study involving human participants was conducted in accordance with the Declaration of Helsinki and applicable institutional and national ethical standards. Ethical approval was obtained from the Ethics Committee of The First Affiliated Hospital of Chongqing Medical University (Approval No. 2025-559-01) and the Ethics Committee of Chong Gang General Hospital (Approval No. 2025-SY-10). Both ethics committees approved a waiver of informed consent because this was a retrospective study. All participant data were anonymized before analysis and handled confidentially to protect individual privacy.
Author Contributions
Congyi Wang (CW): Conceptualization, Methodology, Formal analysis, Writing original draft. Liwen Zhang (LZ): Investigation, Data curation, Validation, Writing - review & editing. Changxi Zhang (CZ): Investigation, Data curation, Writing - review & editing. Gang Li (GL): Validation, Writing - review & editing. Ruiqin Zhou (RZ): Conceptualization, Supervision, Project administration, Writing - review & editing. All authors gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.
Disclosure
The authors report no conflicts of interest in this work.
References
- 1.Chen J, Zhao Y, Yang W. et al. Pulmonary infection after esophageal cancer surgery: impact on the reality, risk factors and development of a predictive nomogram. World J Surg Oncol. 2025;23(1):149. doi: 10.1186/s12957-025-03806-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Chen J, Xiang Q, Zheng XJ, Jiang XY. Predictive model for postoperative pneumonia in patients with esophageal cancer after esophagectomy. Front Oncol. 2025;15:1529308. doi: 10.3389/fonc.2025.1529308 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Huang J, Zhang W, Wang T, et al. Incidence and risk factors of postoperative pulmonary complications following esophageal cancer surgery: a single-center retrospective study. J Thoracic Dis. 2025;17(9):6661–18. doi: 10.21037/jtd-2025-316 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Wang W, Yang T, Xie J, Huang S. Impact of pulmonary infection after radical esophagectomy on serum inflammatory markers, pulmonary function indices, and prognosis. Saudi Med J. 2024;45(1):40–45. doi: 10.15537/smj.2024.45.1.20230504 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Jiang Y, Li Z, Jiang W, Wei T, Chen B. Risk prediction model for postoperative pneumonia in esophageal cancer patients: a systematic review. Front Oncol. 2024;14:1419633. doi: 10.3389/fonc.2024.1419633 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Booka E, Kikuchi H, Hiramatsu Y, Takeuchi H. The Impact of Infectious Complications after Esophagectomy for Esophageal Cancer on Cancer Prognosis and Treatment Strategy. J Clin Med. 2021;10(19):4614. doi: 10.3390/jcm10194614 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Tamagawa A, Aoyama T, Tamagawa H, et al. Influence of Postoperative Pneumonia on Esophageal Cancer Survival and Recurrence. Anticancer Res. 2019;39(5):2671–2678. doi: 10.21873/anticanres.13392 [DOI] [PubMed] [Google Scholar]
- 8.Maruyama S, Shoda K, Kawaguchi Y, et al. Impact of postoperative infectious complications on long-term prognosis after esophagectomy. World J Surg. 2025;49(1):253–261. doi: 10.1002/wjs.12421 [DOI] [PubMed] [Google Scholar]
- 9.Paruk F, Chausse JM. Monitoring the post surgery inflammatory host response. J Emergency Crit Care Med. 2019;3:1. [Google Scholar]
- 10.Tang F, Tie Y, Tu C, Wei X. Surgical trauma-induced immunosuppression in cancer: recent advances and the potential therapies. Clin Transl Med. 2020;10(1):199–223. doi: 10.1002/ctm2.24 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.de Fraiture EJ, Reniers T, Vreeman NEW, et al. Neutrophil phenotypes quantify tissue damage caused by major surgery. Front Surg. 2025;12:1494831. doi: 10.3389/fsurg.2025.1494831 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Austermann J, Roth J, Barczyk-Kahlert K. The Good and the Bad: monocytes’ and Macrophages’ Diverse Functions in Inflammation. Cells. 2022;11(12):1979. doi: 10.3390/cells11121979 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Rao PP, Mishra S, Gupta J, Vyas M, Babu MR. Inflammation and immune biomarkers: new frontiers in understanding and managing diabetes complications. Inflammopharmacology. 2025;33(11):6507–6534. doi: 10.1007/s10787-025-01996-4 [DOI] [PubMed] [Google Scholar]
- 14.Cheng J, Pan Y, Li X, et al. Prognostic Significance of Preoperative Inflammatory and Liver Fibrosis Markers in Hepatocellular Carcinoma. J Inflamm Res. 2025;18:17185–17196. doi: 10.2147/jir.S557334 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Tian T, Xie M, Sun G. Association of systemic immune-inflammation index with asthma and asthma-related events: a cross-sectional NHANES-based study. Front Med. 2024;11:1400484. doi: 10.3389/fmed.2024.1400484 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Islam MM, Satici MO, Eroglu SE. Unraveling the clinical significance and prognostic value of the neutrophil-to-lymphocyte ratio, platelet-to-lymphocyte ratio, systemic immune-inflammation index, systemic inflammation response index, and delta neutrophil index: an extensive literature review. Turkish J Emergency Med. 2024;24(1):8–19. doi: 10.4103/tjem.tjem_198_23 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Wen C, Lan Q, Wang Y, Ni Y, Wong AST, Liu D. Interferon signaling pathways in health and disease. Mol Biomed. 2025;6(1):135. doi: 10.1186/s43556-025-00381-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Qin R. Targeting innate and adaptive immunity to suppress lung cancer metastasis. Front Immunol. 2025;16:1662754. doi: 10.3389/fimmu.2025.1662754 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Le KC, Shuja F, Goronzy JJ, Weyand CM. Immune Age, Cardiovascular Disease, and Anti-Viral Immunity. Cells. 2025;14(22):1793. doi: 10.3390/cells14221793 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Chen Z, Xu Y, Qu C, et al. The IL-33/ST2 Axis Promotes Traumatic Heterotopic Ossification by Driving Macrophage and Mast Cell-Mediated Inflammation via Autophagy Defects. Int J Bio Sci. 2026;22(1):60–85. doi: 10.7150/ijbs.122706 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.An M, Gao F, Zhang B, et al. M2 Macrophage-Derived TGF-β1 Drives Tendon Fibrosis in Rodent Tendinopathy Models. J Orthopaedic Res. 2026;44(1):e70121. doi: 10.1002/jor.70121 [DOI] [PubMed] [Google Scholar]
- 22.Kazemifard N, Golestani N, Jahankhani K, Farmani M, Ghavami SB. Ulcerative colitis: the healing power of macrophages. Tissue Barriers. 2025;13(2):2390218. doi: 10.1080/21688370.2024.2390218 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Hu S, Xu M, Li S, Chen W, Peng Z. LASSO-empowered nomogram integrating nutritional-inflammatory-tumor characteristics predicts immunotherapy outcomes in advanced HCC: large retrospective cohort. Front Immunol. 2025;16:1626940. doi: 10.3389/fimmu.2025.1626940 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Guo R, Zhou X, Chen J, Gao J, Yang Y, Xu K. Clinical significance of inflammation-coagulation-autoantibody triad in systemic sclerosis-associated interstitial lung disease: a retrospective study. Arthritis Res Therapy. 2025;27(1):225. doi: 10.1186/s13075-025-03685-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Xiao J, Chen M, Zhang Y, Tao J. Epigenetic regulation of inflammation dynamics during wound healing: a subtle yet profound shift in histone modifications. Curr Opinion Iimmunol. 2025;96:102635. doi: 10.1016/j.coi.2025.102635 [DOI] [PubMed] [Google Scholar]
- 26.Behmoaras J, Mulder K, Ginhoux F, Petretto E. The spatial and temporal activation of macrophages during fibrosis. Nat Rev Immunol. 2025;25(11):816–830. doi: 10.1038/s41577-025-01186-x [DOI] [PubMed] [Google Scholar]
- 27.Nakamori Y, Park EJ, Shimaoka M. Immune Deregulation in Sepsis and Septic Shock: reversing Immune Paralysis by Targeting PD-1/PD-L1 Pathway. Front Immunol. 2020;11:624279. doi: 10.3389/fimmu.2020.624279 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Margraf A, Ludwig N, Zarbock A, Rossaint J. Systemic Inflammatory Response Syndrome After Surgery: mechanisms and Protection. Anesthesia Analg. 2020;131(6):1693–1707. doi: 10.1213/ane.0000000000005175 [DOI] [PubMed] [Google Scholar]
- 29.Albers-Warlé KI, Helder LS, Groh LA, et al. Postoperative Innate Immune Dysregulation, Proteomic, and Monocyte Epigenomic Changes After Colorectal Surgery: a Substudy of a Randomized Controlled Trial. Anesthesia Analg. 2025;140(1):185–196. doi: 10.1213/ane.0000000000007297 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Jacobs LM, Helder LS, Albers KI, et al. The role of surgical tissue injury and intraoperative sympathetic activation in postoperative immunosuppression after breast-conserving surgery versus mastectomy: a prospective observational study. Breast Cancer Res. 2024;26(1):42. doi: 10.1186/s13058-024-01801-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Frazier WJ, Hall MW. Immunoparalysis and adverse outcomes from critical illness. Pediatr Clin N Am. 2008;55(3):647–68,xi. doi: 10.1016/j.pcl.2008.02.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Li Z, Feng Z, Chen M, et al. Rbfox3 Promotes Transformation of MDSC-Like Tumor Cells to Shape Immunosuppressive Microenvironment. Adv Sci. 2025;12(8):e2404585. doi: 10.1002/advs.202404585 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Li K, Shi H, Zhang B, et al. Myeloid-derived suppressor cells as immunosuppressive regulators and therapeutic targets in cancer. Signal Transduc Target Ther. 2021;6(1):362. doi: 10.1038/s41392-021-00670-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Zhao L, He R, Long H, et al. Late-stage tumors induce anemia and immunosuppressive extramedullary erythroid progenitor cells. Nat Med. 2018;24(10):1536–1544. doi: 10.1038/s41591-018-0205-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Galluzzi L, Buqué A, Kepp O, Zitvogel L, Kroemer G. Immunological Effects of Conventional Chemotherapy and Targeted Anticancer Agents. Cancer Cell. 2015;28(6):690–714. doi: 10.1016/j.ccell.2015.10.012 [DOI] [PubMed] [Google Scholar]
- 36.Spitzer MH, Carmi Y, Reticker-Flynn NE, et al. Systemic Immunity Is Required for Effective Cancer Immunotherapy. Cell. 2017;168(3):487–502.e15. doi: 10.1016/j.cell.2016.12.022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Strauss L, Mahmoud MAA, Weaver JD, et al. Targeted deletion of PD-1 in myeloid cells induces antitumor immunity. Science Immunol. 2020;5(43):1863. doi: 10.1126/sciimmunol.aay1863 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Schlüter T, van Elsas Y, Priem B, Ziogas A, Netea MG. Trained immunity: induction of an inflammatory memory in disease. Cell Res. 2025;35(11):792–802. doi: 10.1038/s41422-025-01171-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Monneret G, Lepape A, Venet F. A dynamic view of mHLA-DR expression in management of severe septic patients. Crit Care. 2011;15(5):198. doi: 10.1186/cc10452 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Volk HD, Reinke P, Krausch D, et al. Monocyte deactivation--rationale for a new therapeutic strategy in sepsis. Intensive Care Med. 1996;22(Suppl 4):S474–81. doi: 10.1007/bf01743727 [DOI] [PubMed] [Google Scholar]
- 41.Gouel-Chéron A, Allaouchiche B, Floccard B, Rimmelé T, Monneret G. Early daily mHLA-DR monitoring predicts forthcoming sepsis in severe trauma patients. Intensive Care Med. 2015;41(12):2229–2230. doi: 10.1007/s00134-015-4045-1 [DOI] [PubMed] [Google Scholar]
- 42.Cheron A, Floccard B, Allaouchiche B, et al. Lack of recovery in monocyte human leukocyte antigen-DR expression is independently associated with the development of sepsis after major trauma. Crit Care. 2010;14(6):R208. doi: 10.1186/cc9331 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Szűcs G, Gézsi A, Szentkereszty M, et al. Human Leukocyte Antigen-DR Expression on Monocytes Is a Useful Predictor in a Systemic Inflammation Response-Based Prognostic Model in Advanced Non-Small Cell Lung Cancer. Int J Mol Sci. 2025;26(18):9226. doi: 10.3390/ijms26189226 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data used to support the findings of this study are available from the corresponding authors upon request.