A Simplified Immune-Dysregulation Index (IL-6/LY) as a Robust Predictor of 28-Day In-Hospital Mortality and MODS in Patients with Sepsis

Introduction

Sepsis is a critical medical condition marked by dysregulated response to infection and subsequent organ dysfunction.¹ Sepsis is associated with a high morbidity and mortality rate despite the advances in intensive care medicine.^2,3 Epidemiological studies have shown that the global annual burden of sepsis is approximately 48.9 million cases and 11 million deaths, accounting for 19.7% of the global death toll. The estimated incidence of sepsis is 189 cases per 100,000 hospitalized patients, with an estimated mortality rate of 26.7%.⁴ A multicenter study across 44 hospitals in China showed a high morbidity rate of sepsis at 20.6% in the intensive care unit (ICU), and the mortality rate ranged from 35.5% to over 50%.⁵ Timely recognition of patients in critical condition enables clinicians to take proactive steps, thus reducing morbidity and mortality. Therefore, identification of effective predictors is a key imperative.

Several clinical scoring systems, such as the sequential organ failure assessment (SOFA) score and acute physiology and chronic health evaluation II (APACHE II) score, are used for risk stratification of patients with sepsis.^6,7 These scoring systems play a crucial role in sepsis management. Their proper application can significantly enhance early sepsis identification, guide interventions effectively, and ultimately lead to improved patient outcomes by quantifying disease severity and predicting patient outcomes.⁸ However, these scores are derived from composite domains that limit their wide application, especially in emergency settings. Therefore, there is a need to develop a more convenient index in real-world practice.

Persistent lymphopenia after the diagnosis of sepsis predicts early and late mortality and may serve as a biomarker for sepsis-induced immunosuppression.⁹ During infection or tissue damage, interleukin-6 (IL-6) is released as a pro-inflammatory cytokine, playing a role in both innate and adaptive immune responses.¹⁰ Previous studies have suggested a prognostic role of IL-6 level in patients with systemic inflammatory response syndrome (SIRS) and multiple organ dysfunction syndrome (MODS). In sepsis, IL-6 may can cause lymphocyte overactivation and apoptosis by promoting inflammation and immune cell activation, and lymphocytopenia may weaken inflammatory control, further intensifying the inflammatory response. Unfortunately, commonly used prognostic biomarkers fail to reflect inflammation and immune status simultaneously in sepsis. For instance, despite extensive research, PCT and lactate mainly indicate bacterial infection and tissue hypoperfusion, not the interplay of inflammation and immune regulation.^11,12 In our earlier study, the IL-6-to-lymphocyte ratio (IL-6/LY) was linked to the prognosis of patients with novel coronavirus disease 2019 (COVID-19) and its prognostic value was superior to that of IL-6 or lymphocyte count alone.¹³ However, the prognostic value of IL-6/LY in patients with sepsis is not clear.

In this study, we investigated the prognostic value of IL-6/LY as a predictor of 28-day in-hospital mortality and incidence of MODS in patients with sepsis.

Materials and Methods

Study Design and Population

This was a single-center, retrospective cohort study. Consecutive adult patients with sepsis who were hospitalized in the Department of Intensive Care Medicine of the First Affiliated Hospital of Soochow University between March 2017 and January 2023 were included. The follow-up time for assessing mortality outcomes was 28 days after admission to the ICU. The study protocol conformed to the Declaration of Helsinki and was approved by the Ethics Committee of the First Affiliated Hospital of Soochow University (IRB No. 2021–274).

Inclusion and Exclusion Criteria

Sepsis is diagnosed based on the Sepsis 3.0 criteria, which requires the presence of evidence of infection along with a SOFA score of ≥2.¹ Only adult patients (age ≥18 years) for whom data regarding IL-6 levels, lymphocyte counts, and outcome measurements were available were eligible for inclusion. Patients with hematological disorders affecting lymphocyte counts, those recently treated with immunosuppressants, or those receiving any chemoradiotherapy were excluded.

Clinical Data Collection

The baseline characteristics of the patients, including age, sex, body mass index (BMI), comorbidities, primary infection site, and pathogen, as well as APACHE II and SOFA scores, were recorded. These scores were routinely calculated at our hospital during the study reference period. Data regarding the ICU length of stay (LOS), invasive mechanical ventilation, and continuous renal replacement therapy (CRRT) were also collected. We retrospectively collected the results of routine blood tests, serum cytokine measurements, blood coagulation function tests, biochemistry, and arterial blood gas analysis performed within 24h of admission were collected from the hospital database. IL-6/LY was calculated by dividing IL-6 (pg/mL) by the lymphocyte count (×10⁹/L).

IL-6 Test

Sepsis patients at our hospital were routinely tested for cytokines using a 12-in-1 cytokine detection kit (multiple microsphere flow immunofluorescence method). The BD Canto II flow cytometer was used to measure the IL-6 level.

Follow Up

The primary outcome of this study was the 28-day in-hospital mortality after ICU admission. The secondary outcome was the incidence of MODS. MODS is defined as acute and potentially reversible dysfunction or failure of two or more organ systems,^14,15 including the cardiovascular, hepatic,¹⁶ coagulation,¹⁷ respiratory,¹⁸ renal,¹⁹ and other organs,¹⁵ due to severe conditions like serious infections.

Statistical Analysis

All continuous variables were non-normally distributed and were presented as median (lower quartile, upper quartile). Comparisons were made using the Mann–Whitney U-test. Categorical variables were presented as frequency (percentage) and subjected to comparison through the chi-squared test or Fisher’s exact test, as deemed suitable Patients were categorized into high IL-6/LY and low IL-6/LY groups based on a cutoff value determined by the Youden index. Multivariate COX and logistic regression models were employed to assess the association of IL-6/LY with overall survival and MODS. Restricted cubic splines were utilized to illustrate a nonlinear correlation between IL-6/LY and mortality. ROC curve analysis was conducted to evaluate the prognostic value of IL-6/LY. Survival analysis was executed using the Kaplan‒Meier method, and differences in survival between groups were evaluated using the Log rank test. Additionally, a subgroup analysis was carried out using COX and logistic regression, along with an examination of the interaction between the given stratification and the study group. P values < 0.05 were considered indicative of statistical significance. Statistical analyses were performed using the R software (version 4.0.3, Vienna, Austria).

Results

Baseline Characteristics

Figure 1 displays a schematic representation of the study design and criteria used for patient selection. A total of 782 patients with sepsis qualified the diagnostic criteria for sepsis described above during the study reference period. Based on the exclusion criteria, 301 patients were included in this study. Of these, 122 (40.5%) patients were in the high IL-6/LY group and 179 (59.5%) patients in the low IL-6/LY group according to the cutoff value of 326.04 (Youden index 0.813; sensitivity: 83%; specificity: 84%).

Figure 1 Flow chart showing the study design and patient-selection criteria.

The baseline patient characteristics are presented in Table 1. Compared to those with low IL-6/LY, patients with high IL-6/LY were older, had faster heart rate and respiratory rate, and had higher APACHE II and SOFA scores (P < 0.050, respectively). There were no significant between-group differences with respect to sex, BMI, blood pressure, or history of comorbidities. The distribution of sites of infection and pathogen types was comparable between the two groups. Patients with high IL-6/LY had significantly higher levels of lactic acid, IL-6, NT-proBNP, PT, APTT, creatinine, and AST, and significantly lower lymphocyte count, platelet count, WBC count, PH, PO₂, and oxygenation index compared to those with low IL-6/LY (P < 0.050 for all). Other laboratory parameters were not significantly different between the two groups. Compared to the low IL-6/LY group, the high IL-6/LY group showed a shorter length of free-CRRT time, length of free-mechanical ventilation time, and LOS in the ICU, while the duration of vasoactive drug use was prolonged (P < 0.050 for all). In addition, 28-day in-hospital mortality and incidence of MODS in the high IL-6/LY group were significantly higher than that in the low IL-6/LY group (P < 0.001 for both).

Table 1 Baseline Characteristics of Septic Patients in Low and High IL-6/LY Groups

Association Between IL-6/LY and 28-Day In-Hospital Mortality

Multivariate COX models were employed to evaluate the correlation between IL-6/LY and the risk of 28-day in-hospital mortality (Table 2). In the unadjusted model, high IL-6/LY was associated with a 10-fold higher risk of death (hazard ratio [HR]: 10.30, 95% CI 6.27–16.94, P < 0.001). After adjusting for age, sex, hypertension, diabetes, COPD, lactic acid, platelet count, PT, AST at admission, serum creatinine at admission, and SOFA score, IL-6/LY remained a significant predictor of 28-day in-hospital mortality (HR: 8.01, 95% CI 4.67–13.74, P < 0.001).

Table 2 Multivariate Cox Regression Analysis for the Association Between IL-6/LY and 28-Day in-Hospital Mortality

We used a restricted cubic spline to analyze the shape of the association between IL-6/LY and 28-day in-hospital mortality. The results showed a positive relationship between the IL-6/LY and 28-day in-hospital mortality (P_{Non-linearity} < 0.001) (Figure 2). The Kaplan‒Meier survival curves demonstrated a significantly higher risk of 28-day in-hospital mortality in the high IL-6/LY group compared to that in the low IL-6/LY group (P < 0.001) (Figure 3). To further analyze the role of IL-6/LY in different populations, a subgroup analysis was performed. Results of subgroup analysis stratified by patient age, sex, history of hypertension, and diabetes are presented in Figure 4. In this sensitivity analysis, the IL-6/LY remained a significant predictor of mortality, and there was no difference in trends between subgroups (P < 0.050 for all; P-interaction > 0.050). As shown in Figure 5, IL-6/LY (AUC = 0.893, 95% CI, 0.855–0.931, P < 0.001) outperformed the SOFA score (AUC = 0.717, 95% CI, 0.655–0.778, P < 0.001) in predicting 28-day in-hospital mortality.

Figure 2 Restricted cubic spline analysis of the relationship between IL-6/LY and 28-day in-hospital mortality in patients with sepsis. Dashed lines are 95% confidence intervals (P non-linearity < 0.001).

Figure 3 Kaplan‒Meier cumulative survival curves for 28-day in-hospital mortality in patients with sepsis according to IL-6/LY. (log-rank P < 0.001).

Figure 4 Subgroup analysis of the association between IL-6/LY and 28-day in-hospital mortality. Abbreviation: HR, hazard ratio.

Figure 5 The ROC curves of SOFA and IL-6/LY to predict the 28-day in-hospital mortality in sepsis patients (AUC = 0.717, 95% CI, 0.655–0.778; AUC = 0.893, 95% CI, 0.855–0.931, respectively).

Association Between IL-6/LY and MODS

Multivariate logistic analysis was conducted to evaluate the correlation between IL-6/LY and the risk of developing MODS. As shown in Table 3, both univariate (OR 5.62, 95% CI 3.36‒9.43, P < 0.001) and multivariate (OR 3.44, 95% CI 1.85‒6.38, P < 0.001) analyses showed that a higher IL-6/LY was associated with an increased risk of MODS.

Table 3 Predictors of MODS in Univariable and Multivariable Logistic Regression Analyses

As shown in Figures 6 and 7, patients with MODS had higher IL-6/LY levels than those without MODS, and the incidence of MODS was higher in the high IL-6/LY group than in the low IL-6/LY group. Subgroup analysis of the role of IL-6/LY in predicting MODS is shown in a stratified forest plot in Figure 8. The results showed that IL-6/LY remained a significant risk factor for MODS, except in patients with diabetes. The ROC curves for SOFA score and IL-6/LY for predicting the risk of MODS are shown in Figure 9. SOFA score and IL-6/LY showed similar predictive efficacy for the incidence of MODS (AUC = 0.740, 95% CI, 0.685–0.795; AUC = 0.743, 95% CI, 0.688–0.798, respectively).

Figure 6 IL-6/LY levels in MODS and non-MODS patients. ***P < 0.001.

Figure 7 Bar graph of MODS incidence in low and high IL-6/LY groups.

Figure 8 Subgroup analyses of the association between IL-6/LY and MODS. Abbreviation: OR, odds ratio.

Figure 9 The ROC curves of SOFA and IL-6/LY in predicting MODS (AUC = 0.740, 95% CI, 0.685–0.795; AUC = 0.743, 95% CI, 0.688–0.798, respectively).

Discussion

In this retrospective cohort study, a high IL-6/LY in patients with sepsis was associated with an increased risk of 28-day in-hospital mortality and MODS. Following adjustments for demographic and clinical characteristics in the multivariate model, IL-6/LY retained its significance as a predictor for adverse outcomes. Patients in the high IL-6/LY group had more unstable vital signs, poor survival, and a high incidence of organ dysfunction. Subgroup analysis revealed the prognostic value of IL-6/LY in most patients with sepsis.

Cytokine storm and immunosuppression are two key drivers of the development and progression of sepsis, contributing to increased morbidity and mortality.²⁰ Research conducted in the past decade has unraveled the critical role of immune function and inflammatory response in the development of sepsis.^21,22 IL-6 is induced by stress, including infection and trauma. It plays an essential role in host defense by stimulating acute-phase responses, hematopoiesis, and immune reactions.²³ Nevertheless, exaggerated production of IL-6 predisposes to cytokine storm,¹⁰ a pernicious reaction associated with an adverse prognosis. Song et al established the diagnostic and prognostic values of IL-6 in individuals diagnosed with sepsis.¹¹ In a prospective multicenter observational study, elevated IL-6 expression was identified as being correlated with mortality in ICU patients with sepsis,²⁴ which is consistent with our findings. Cifaldi et al reported that elevated levels of IL-6 were associated with impaired cytolysis due to hyperactivation of the immune system, resulting in MODS.²⁵ In our study, IL-6/LY showed better predictive efficacy and was equally statistically significant in logistic and COX regression and subgroup analysis.

Immunosuppression induced by sepsis is characterized by the depletion of apoptosis-associated immune cells, including CD4⁺ and CD8⁺ T lymphocytes, B lymphocytes, and follicular dendritic cells.²⁶ Persistent lymphopenia, a marker of sepsis-induced immunosuppression, is considered a predictor of mortality.⁹ Compared to survivors, deceased patients with sepsis showed a lower quantity of mature B cells and circulating Tfh (cTfh) cells. Lymphocyte counts are inversely associated with mortality.²⁷ Additionally, dysregulated lymphocyte function was shown to be associated with the occurrence and prognosis of MODS in patients with sepsis.²⁸ Humanized mouse models of sepsis suggest that the decrease in septic lymphocytes may be related to their regulation.²⁹ Overall, in sepsis, IL-6 may promote inflammatory responses and cause immune cell overactivation and tissue damage by activating multiple signaling pathways, such as the JAK/STAT3 pathway.³⁰ This novel indicator, which combines inflammatory responses and lymphocyte regulation, can more comprehensively reflect the stress response during sepsis. To the best of our knowledge, this study represents the initial exploration of the prognostic significance of IL-6/LY in sepsis.

Multi-dimensional scoring systems, including the APACHE II and SOFA scores, are well-known risk stratification tools for sepsis management. However, these models rely on an array of demographic, clinical, and laboratory data, rendering them inconvenient to implement in emergency settings. This significantly limits their application in routine clinical practice, especially for more intensive scenarios, typically in the emergency department.^31,32 In contrast, the predictive efficacy of IL-6/LY is similar to that of SOFA score, but it employs two simple serum biomarkers, making it more convenient to use in this setting. In addition, IL-6 assays have gradually become routine and more readily available in clinical practice, thereby making IL-6/LY a more appropriate method for point-of-care applications. Therefore, IL-6/LY facilitates more timely sepsis monitoring and earlier intervention, enhancing the potential for improved patient outcomes.

Compared with the low IL-6/LY group, WBC count was lower in the high IL-6/LY group, which is a perplexing finding. This might be related to the intensity of antibiotic use and bone marrow suppression caused by severe infection.^33,34 Moreover, the high IL-6/LY group had a shorter ICU LOS, likely due to the severity of the condition. Relevant studies have shown that the mortality rate of septic patients is high in the early stage.³⁵ Some limitations of this study should be considered while interpreting the results. Firstly, it was performed in a single tertiary center and the study population comprised exclusively of Han Chinese patients. Secondly, the study’s retrospective nature precludes causal inferences and may leave some confounding factors unavoidable. Finally, the follow-up period for the survival analysis was confined to 28 days, and further study is required to assess the long-term prognostic value. A prospective multicenter cohort with a longer follow-up duration is currently underway to address these limitations.

Conclusion

The IL-6/LY may become a novel prognostic biomarker for predicting 28-day in-hospital mortality and MODS in patients with sepsis. This composite index integrates IL-6 and lymphocyte counts, reflecting the dual pathological drivers of sepsis: hyperinflammation and immune paralysis. Elevated IL-6/LY values correlate with systemic inflammation, impaired host defense, and increased risk of organ failure, offering a more holistic assessment than traditional biomarkers. Its dual-parameter integration enhances prognostic accuracy for early identification of high-risk patients and may guide real-time therapeutic decisions, with potential for future validation in personalized critical care management.