Predictive value of elevated interleukin-33 levels for multi-organ dysfunction syndrome in trauma patients in South Korea: a prospective observational study

Sanghyun An; In Sik Shin; Myoung Jun Kim; Da Kyung Kim; Md Habibur Rahman; Cheol-Su Kim; Kwangmin Kim

doi:10.4266/acc.002500

Articles

Page Path: HOME > Acute Crit Care > Volume 40(4); 2025 > Article

Original Article Trauma Predictive value of elevated interleukin-33 levels for multi-organ dysfunction syndrome in trauma patients in South Korea: a prospective observational study: Acute and Critical Care 2025;40(4):594-604.
DOI: https://doi.org/10.4266/acc.002500
Published online: November 28, 2025

Sanghyun An^1,2

, In Sik Shin³

, Myoung Jun Kim⁴

, Da Kyung Kim⁵

, Md Habibur Rahman⁶

, Cheol-Su Kim⁶

, Kwangmin Kim^7,8

¹Division of Colorectal Surgery, Department of Surgery, Yonsei University Wonju College of Medicine, Wonju, Korea

²Center of Evidence Based Medicine, Institute of Convergence Science, Yonsei University, Seoul, Korea

³Division of Acute Care Surgery, Department of Surgery, Yonsei University Wonju College of Medicine, Wonju, Korea

⁴Department of Trauma Surgery, Armed Forces Capital Hospital, Seongnam, Korea

⁵Yonsei University Wonju College of Medicine, Wonju, Korea

⁶Department of Convergence Medicine, Yonsei University Wonju College of Medicine, Wonju, Korea

⁷Health Check-up Center, Wonju Severance Christian Hospital, Wonju, Korea

⁸Institute of Evidence Based Medicine, Yonsei University Wonju College of Medicine, Wonju, Korea

Corresponding author: Kwangmin Kim Health Check-up Center, Wonju Severance Christian Hospital, 20 Ilsan-ro, Wonju 26426, Korea Tel: +82-33-741-0570Fax: +82-33-741-0574 E-mail: lukelike@yonsei.ac.kr

• Received: July 13, 2025 • Revised: September 1, 2025 • Accepted: September 4, 2025

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

785 Views
68 Download

prev next

Full Article

Download PDF

Abstract
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
KEY MESSAGES
NOTES
SUPPLEMENTARY MATERIALS
REFERENCES

Abstract

Background
Multi-organ distress syndrome (MODS) causes morbidity in patients with trauma. This study evaluates the effectiveness of interleukin-33 (IL-33), which reflects tissue damage and the inflammatory response, as a MODS indicator in patients with trauma.
Methods
Patients with trauma admitted to our trauma center between July 2022 and July 2023 were included. IL-33 levels were measured in blood samples for 4 days. Correlations with clinical and laboratory indicators, including initial IL-33 levels, were analyzed to identify independent predictors of MODS.
Results
Among the 87 patients enrolled, 20 developed MODS. Initial IL-33 levels were elevated in the MODS group, compared with the non-MODS group. In the non-MODS group, IL-33 levels increased on day 1 and then declined, whereas in the MODS group, IL-33 levels were highest at admission (day 0) and decreased continuously through day 3. In patients with detectable initial IL-33 levels, the measured levels correlated with higher Abbreviated Injury Scale 5 scores and the Injury Severity Score (ISS). A logistic regression analysis revealed the ISS and delta neutrophil index as factors contributing to MODS progression.
Conclusions
The findings suggest that initial IL-33 levels are elevated in the MODS group, compared with non-MODS group, and exhibit a rapidly declining trend, showing an initial association with MODS that was not maintained in a multivariate analysis. These findings suggest that IL-33 might have relevance in assessing trauma severity; however, further validation is required before it can be considered a biomarker for MODS.
Key Words: biomarkers; interleukins; multi-organ distress syndrome; trauma

INTRODUCTION

Severe traumatic injury is associated with high mortality and morbidity rates [1]. Recent advances in therapeutic modalities targeting hemorrhage control and coagulopathy management, combined with the implementation of an integrative, interdisciplinary approach to the care of patients with trauma, have diminished the incidence of immediate fatalities. However, the persistence of secondary complications, particularly multi-organ distress syndrome (MODS), remains a significant and enduring cause of morbidity among patients hospitalized for trauma [2,3].

The immune response following tissue damage can be triggered by alarmins, a diverse set of molecules constitutively expressed and released when cellular damage occurs [4]. Because they are likely to be the primary triggers of the immune response following injury, there is considerable scientific curiosity about their biology. Moreover, their prospective utility as biomarkers or therapeutic targets across various conditions, notably in addressing inflammatory responses after significant trauma, is currently gaining attention.

Interleukin-33 (IL-33) is a member of the IL-1 cytokine family and functions as an alarm signal in the immune system [5]. The complete IL-33 protein requires enzymatic cleavage by proteases to produce its mature biologically active state. This active form can then bind to its specific receptor, the suppression of tumorigenicity (ST2) receptor, and initiate the activation of the nuclear factor kappa B (NF-κB) pathway across diverse innate and adaptive immune cell types. An elevated level of IL-33, primarily within epithelial and endothelial cells, contributes to its proinflammatory functions [6].

Diverse biomarkers are used to anticipate the adverse outcomes associated with significant inflammation in critical ill individuals. These indicators encompass parameters such as white blood cell (WBC) count and concentrations of lactic acid, procalcitonin, and C-reactive protein (CRP) [7,8]. Recently, research on molecular biomarkers, including mitochondrial DNA, toll-like receptors, and cytokines, has aimed to clarify their value in the early identification of adverse outcomes in critically ill patients [9-11]. Unexpectedly, despite evidence of its role in other inflammatory conditions, IL-33 has rarely been investigated as a predictor in trauma patients. Additionally, studies on the characteristics of IL-33 and its correlation with other clinical factors in trauma cases remain inadequate.

Therefore, our primary aim in this study was to evaluate the predictive value of IL-33 as an indicator of the onset of MODS in trauma patients. Furthermore, we sought to explore the association between IL-33 levels and diverse clinical features within this particular patient cohort.

MATERIALS AND METHODS

This single-center prospective observational study was conducted after approval from the Ethics Committee of Wonju Severance Christian Hospital (No. CR322047). Written informed consent was obtained from the patients. The study protocol for this research was registered at clinicaltrials.gov under the identifier NCT05441787.

Patient Selection

Trauma patients who were admitted to the department of trauma surgery via our trauma center were recruited from July 2022 to July 2023. The exclusion criteria were as follows: (1) individuals younger than 18 years, (2) pregnant individuals, (3) patients presenting with fatal conditions upon initial assessment, (4) patients who declined to provide informed consent, and (5) those without a legal representative or with severely impaired consciousness that prevented informed consent. After applying the exclusion criteria, 87 patients were enrolled in this study (Figure 1).

Data Collection and Definition

After obtaining written informed consent, we collected pertinent clinical background information and assessed for the presence of additional comorbidities. The following information was prospectively collected: age, sex, medical history, mechanism of trauma, systolic blood pressure (SBP), Injury Severity Score (ISS), Abbreviated Injury Scale 1–6 (AIS1–6), and laboratory parameters at initial presentation (delta neutrophil index [DNI], WBC count, neutrophil count, creatinine level, hemoglobin level, platelet count, international normalized ratio [INR], CRP level, lactate level, and IL-33 level). Additionally, IL-33 levels were measured on the first, second, and third day after admission to monitor daily changes. Undetectable IL-33 values (below the assay’s detection limit) on any measurement day (baseline, days 1–3) were replaced with 0, as a conservative approach to retain all patients in the analysis. The Sequential Organ Failure Assessment (SOFA) score was systematically computed and documented throughout each patient’s intensive care unit (ICU) stay. Following patient discharge, we recorded the number of packed red blood cell (RBC) transfusions, length of stay (LOS), ICU LOS, MODS occurrence, and mortality. MODS was determined as a SOFA score of 6 or higher on 2 consecutive days that occurred at least 48 hours post-admission to the emergency department [12].

Measurement of IL-33 Levels

Blood samples were collected using BD Vacutainers to allow clotting. The samples were transferred to the central laboratory within 2 hours of collection. Subsequently, centrifugation was performed for 10 minutes at ×g 1,000. Following centrifugation, the specimens were preserved at a temperature –80 °C until they underwent multiplex analysis to detect cytokines. After the samples were dissolved, the multiplex analysis was performed using a Bio-Rad multiplex bead array system (Bio-Rad). Bio-Plex assays (Bio-Rad) are equipped with standard concentrations of each analyte to facilitate accurate determination of the concentrations of the proteins of interest against calculated standard curves. All procedures were conducted according to the manufacturer's guidelines. In summary, lyophilized cytokine standards were reconstituted using standard diluents, and serial dilution series involving 30 μL of standard were executed to establish standard curves for each immune regulator under investigation. The bead mixture, designed specifically for immune molecules, was introduced to 30 μl of either standard or serum samples. For serum samples, a 1:2 dilution was made using sample diluent. Next, the plate underwent an overnight incubation of 16 to 18 hours at a temperature of 4 °C. To ensure proper cleaning and preparation, a series of washing steps was conducted, each involving the application of 100 μl of wash buffer per well and facilitated by an automated washer specifically designed for magnetic beads. Next, the detection antibody was introduced to the plate and allowed to incubate for 1 hour at room temperature. Subsequently, a mixture containing streptavidin-phycoerythrin was added and incubated for an additional 30 minutes at room temperature. Upon completion of those incubation periods, three final washing steps were performed to ensure proper preparation. The beads were then resuspended using 125 μL of assay buffer sourced from Bio-Rad. Finally, the analysis was conducted using a Luminex 200 Bio-Plex instrument from Bio-Rad.

Statistical Analyses

Statistical analyses were conducted in R statistical software (version 4.1.0; R Foundation for Statistical Computing). Continuous variables are depicted as either means accompanied by standard deviations or median values with ranges. To assess differences among the variables, we used an independent sample t-test for normally distributed data and the Mann-Whitney U-test for non-normally distributed data. Comparative analyses of categorical variables were conducted using the chi-square test and Fisher’s exact test. To identify independent risk factors, we conducted a multivariate logistic regression analysis. To identify independent risk factors in more severely injured patients, the same statistical methods were used for a subgroup of patients with an ISS ≥16. Graphs were constructed using GraphPad Prism (GraphPad Software). Statistical significance was defined as P<0.05.

RESULTS

Comparison of Enrolled Trauma Patients with and without MODS

A significant difference between the groups was observed in their mean age, 55.6±16.7 years in patients without MODS and 64.0±16.8 years in those with MODS (P=0.035). The difference in the proportion of males (patients without vs. with MODS: 56 [83.6%] vs. 15 [75.0%], P=0.511) was not significant between the groups, but SBP did differ significantly between the groups, with values of 122.5±33.8 and 95.3±32.5, respectively (P=0.001). The ISS (14.6±8.1 vs. 26.0±10.2, P<0.001) and AIS1 (0.5±1.1 vs. 1.4±1.6, P=0.014), AIS3 (1.6±1.5 vs. 2.5±1.5, P=0.023), and AIS5 (1.0±1.4 vs. 2.0±1.9, P=0.037) scores were significantly lower in patients without MODS than in those with MODS. The initial DNI scores (patients without vs. with MODS: 1.9±2.0 vs. 4.5±3.9, P=0.003), creatinine levels (1.0±1.0 vs. 1.4±0.6, P<0.001), hemoglobin levels (12.4±2.0 vs. 11.3±2.3, P=0.025), INR (1.1±0.2 vs. 1.2±0.2, P<0.001), and lactate levels (2.7±2.0 vs. 5.7±4.0, P<0.001) differed significantly between the groups. Patients with MODS exhibited significantly higher worst SOFA scores than those without MODS (7.8±2.1 vs. 1.9±1.8, respectively, P<0.001). There was a considerable disparity in the number of packed RBC transfusions administered within a 24-hour period, with patients without MODS receiving significantly lower volumes than those with MODS (1.6±2.9 vs. 6.3±6.6, respectively, P<0.001). Patients with MODS experienced notably extended durations of hospitalization compared with those without MODS, with a hospital LOS of 34.8±27.9 days vs. 18.6±18.5 days, respectively, (P=0.016). Similarly, ICU stays were significantly prolonged in patient with MODS, 11.7±9.8 days, compared with 4.9±8.3 days in patients without MODS (P<0.001). The incidence of mortality was notably higher among patients with MODS than in those without MODS, with mortality rates of 3 [15.0%] vs. 0 [0%], respectively (P=0.011) (Table 1).

Differences in Patients with ISS ≥16 with and without MODS

The mean age (57.8±16.5 vs. 64.8±17.6, P=0.112) and male sex ratio (29 [87.9%] vs. 13 [72.2%], P=0.249) did not differ significantly between patients without and with MODS. However, SBP did differ significantly different between the groups, with values of 113.6±38.6 vs. 94.9±33.2, respectively (P=0.039). The ISS was notably higher in patients with MODS than in those without (27.8±8.9 vs. 21.2±5.6, respectively, P=0.001). The initial DNI scores (patients without vs. with MODS: 2.2±2.5 vs. 4.2±3.4, P=0.036), creatinine levels (1.2±1.4 vs. 1.4±0.7, P=0.014), INR (1.1±0.2 vs. 1.2±0.2, P=0.017), and lactate levels (3.3±2.5 vs. 6.1±4.1, P=0.003) differed significantly between the groups. Moreover, compared with those without MODS, patients with MODS exhibited significantly higher worst SOFA scores (2.6±1.9 vs. 7.9±2.2, respectively, P<0.001). The number of packed RBC transfusions within 24 hours was significantly higher in patients with MODS than in those without the syndrome (7.0±6.5 vs. 2.6±3.8, respectively, P=0.007). Additionally, the ICU LOS was significantly longer in patients with MODS than in those without MODS (12.4±10.1 vs. 5.5±5.3, respectively, P=0.008). The incidence of mortality was significantly higher among patients with MODS than among those without MODS (3 [16.7%] vs. 0 [0%], respectively, P=0.039) (Table 1).

Daily Changes in IL-33 Levels

The IL-33 levels in all patients from day 0 to day 3 are illustrated in Supplementary Figure 1. The IL-33 levels in all patients who did not develop MODS increased until day 1 and then decreased. In contrast, the IL-33 levels in all patients who developed MODS consistently decreased from day 0 to 3. The initial IL-33 levels were significantly higher in patients with MODS than in those without the syndrome (12.9±38.6 vs. 4.0±13.8, respectively, P=0.004).

In patients with an ISS ≥16 who did not develop MODS, IL-33 levels increased until day 2 and then decreased. However, in patients with an ISS ≥16 who developed MODS, IL-33 levels decreased consistently from day 0 to 3. The initial concentration of IL-33 in patients diagnosed with MODS was significantly higher than in individuals without MODS (13.9±40.6 vs. 3.4±11.3, respectively, P=0.019) (Table 2).

Predictors of MODS in Trauma Patients

To identify independent risk factors in all enrolled patients, logistic regression models encompassing the following variables were used: age; sex; SBP; ISS; AIS1, AIS3, and AIS5 scores; IL-33 level; DNI scores; creatinine and hemoglobin levels; INR; and lactate level. Only the ISS (odds ratio [OR], 1.13; 95% CI, 1.05–1.21; P<0.001) and DNI scores (OR, 1.35; 95% CI, 1.09–1.67; P=0.006) exhibited independent correlations with MODS.

Similarly, we used a logistic regression model to identify the independent risk factors in patients with an ISS ≥16. That model included age, sex, SBP, IL-33 level, DNI, creatinine level, INR, and lactate level. Among them, only lactate levels were independently correlated with MODS (OR, 1.36; 95% CI, 1.09–1.69; P=0.006) (Table 3).

Correlation between IL-33 Levels and Clinical Characteristics

The daily levels of IL-33, with their means and standard deviations, are shown in Figure 2. Of all the enrolled patients, 44 (50.6%) had IL-33 levels below the detection threshold at initial presentation. Consequently, we stratified the entire cohort of enrolled participants into two subgroups: one comprising individuals with detectable IL-33 levels (IL-33P group) and the other consisting of those with undetectable IL-33 levels (IL-33N group). Subsequently, we performed a comprehensive analysis of their clinical characteristics, particularly their ISS, AIS scores, various laboratory findings, and the number of packed RBC transfusions within 24 hours.

The ISS (14.2±7.9 vs. 20.3±10.7, P=0.003) and AIS5 score (0.9±1.2 vs. 1.6±1.8, P=0.044) were significantly lower in the IL-33N group than in the IL-33P group. The WBC count (14.0±5.4 vs. 14.7±7.4, P=0.608), neutrophil count (11.8±4.9 vs. 11.9±7.0, P=0.913), DNI (2.3±2.8 vs. 2.7±2.8, P=0.442), INR (1.1±0.2 vs. 1.1±0.2, P=0.795), and number of packed RBC transfusions (1.8±3.8 vs. 3.5±5.0, P=0.076) did not differ significantly between the groups (Figure 3).

DISCUSSION

In our cohort, 20 patients (23.0%) developed MODS. This relatively high incidence can be attributed to the fact that our institution is a regional trauma center for a wide geographic area, which results in a high proportion of patients with severe trauma. In our cohort, 51 patients (58.6%) had an ISS of ≥16, of whom 18 developed MODS, indicating that the substantial proportion of severely injured patients likely contributed to the observed incidence of MODS.

Our prospective investigation unveiled IL-33 as a promising biomarker for anticipating the onset of MODS in trauma patients. Although IL-33 was not an independent predictor in our logistic regression analysis, it was included in the final regression model in patients with ISS ≥16. Divergent IL-33 levels were observed between patients who did and did not develop MODS. Human IL-33 is predominantly detected in endothelial cells that line the vascular system and epithelial cells at particular barrier sites. It exhibits diverse levels of constitutive expression across various stromal cell populations [13,14]. The precise role of its robust constitutive vascular expression patterns in quiescent endothelial cells remains unclear. However, this localization suggests that IL-33 is strategically involved in a prompt systemic immune response following endothelial injury. The interaction between released IL-33 and its receptor ST2 enables ST2 to bind with the IL-1 receptor accessory protein, which acts as a coreceptor within the IL-1 superfamily. This IL-33 receptor complex initiates signaling cascades that involve myeloid differentiation primary response 88, IL-1 receptor-associated kinase (IRAK) 1, and IRAK4 kinases, as well as tumor necrosis factor receptor-associated factor 6. These signaling pathways ultimately lead to the activation of various mitogen-activated protein kinase kinases and the NF-κB pathway, which in turn lead to a proinflammatory response. Therefore, increased IL-33 levels could be associated with severe inflammatory reactions such as MODS.

However, our results appear to contradict those of previous studies. Krychtiuk et al. [15] reported that reduced levels of IL-33 and elevated levels of ST2 predicted mortality in critically ill patients. Moreover, Billiar et al. [16] linked reduced IL-33 levels coupled with increased ST2 receptor levels with unfavorable outcomes in individuals experiencing blunt trauma. They explained that after IL-33 bound to ST2, the complex played a positive role in tissue repair. That discovery is in line with the function of IL-33 in stimulating both local and systemic type 2 responses, which are essential for establishing homeostasis and facilitating repair processes after significant trauma. Therefore, an initial surge in IL-33 levels indicates a greater possibility of binding to ST2 and is associated with better outcomes in patients with trauma. The observed discrepancy in the effects of IL-33 in humans could stem from the highly pleiotropic nature of this cytokine. It exhibits both proinflammatory effects on innate immune cells and anti-inflammatory effects on T regulatory cells, making it challenging to anticipate the overall impact of an IL-33 surge. Interestingly, we observed an IL-33 surge at initial presentation followed by a rapid decrease in patients with MODS, regardless of whether their ISS was ≥16. After day 1, IL-33 levels were lower in patients with MODS than in those without MODS. Therefore, the net effect of the IL-33 surge at initial presentation might be proinflammatory, and IL-33 levels maintained after 1 day might have stronger anti-inflammatory effects on T regulatory cells and play a role in preventing MODS. Another reason for this discrepancy could be the blood sampling time. In Krychtiuk et al. ’s study [15], blood was drawn from an arterial or central venous line within 24 hours of admission. However, we analyzed an initial blood sample taken at the initial presentation of the enrolled patients, and we found that the initial level of IL-33 was the highest. Sundnes et al. [17] also reported a more rapid surge in IL-33 levels, within 8 hours after injury, in critically injured patients (new ISS >24) than in less injured patients. Therefore, that IL-33 surge might not have been detected in the study by Krychtiuk et al. [15] because of differences in sampling time. The differences in results between studies could also be due to differences in pre-analytical conditions, analysis methods, and the baseline characteristics of the enrolled patients, such as age, injury severity, and mechanisms.

The ISS is an anatomical scoring system designed to comprehensively assess patients with multiple injuries. It involves squaring the scores of the three most severely injured body regions and then summing them to derive the ISS value. In other words, the ISS reflects severe localized injury and the extent of injury. As previously mentioned, IL-33 originates from endothelial cells lining the vascular system and epithelial cells at specific barrier surfaces, and it exhibits diverse levels of constitutive expression in various stromal cell types. A large extent of injury causes extensive damage to those cells. In line with that understanding, the ISS was higher in the group that tested positive for IL-33 than in the group without detectable IL-33. In addition, we assessed the effects of various anatomical injury patterns. No significant variance was found between the IL-33-positive and IL-33-negative groups across AIS1, 2, 3, 4, and 6. However, a noticeable difference was observed at AIS5, “extremities or pelvic girdle,” which represents the broadest section among AIS1–6.

Our investigation revealed serum lactate levels as an autonomous predictor of MODS progression among trauma patients with ISS ≥16. The serum lactate level is a biomarker commonly used in critically ill individuals because it reflects the extent of tissue hypoxia and anaerobic metabolism [18]. This observation aligns with previous research indicating that elevated lactate levels correlate with compromised oxygen delivery and utilization, both pivotal factors in the pathophysiology of MODS [19]. In circumstances such as sepsis, shock, and respiratory failure, the oxygen supply to tissues is inadequate, necessitating reliance on anaerobic metabolism, which generates lactate as a byproduct. Normally cleared by the liver and kidneys, lactate can accumulate when those organs are over-saturated or when production exceeds the clearance rate, signaling metabolic distress and potentially indicating a high risk of conditions such as MODS [20].

The DNI has been reported to be a useful biomarker for predicting disease severity and outcomes in patients with infections, sepsis, and severe injuries [21,22]. In our study, the DNI was also associated with the development of MODS. IL-33 is a Damage associated molecular pattern (DAMP) that can activate innate immune responses through the production of proinflammatory cytokines [23]. Consequently, we hypothesized that IL-33 would serve as a more rapid and precise biomarker for predicting MODS than other DAMP factors, such as the DNI, or laboratory markers of anaerobic metabolism, such as lactate levels. Contrary to our expectations, our findings do not align with that hypothesis. This discrepancy in the expected outcomes could be attributed to the timing of sample collection and the multifaceted role of IL-33, which might have influenced the observed results. If IL-33 measurements were made immediately after injury, the findings could yield divergent results. Therefore, expediting patient transfer might enhance the clarity of IL-33 as a biomarker. However, given our institution’s status as a level I trauma center and the effectiveness of our patient transport system, lactate level and DNI have emerged as the primary biomarkers for predicting MODS in patients with trauma.

Our study has certain limitations. First, the small sample size and limited representation of patients with MODS might not fully capture the diversity of the broader patient population. Second, the wide standard deviation observed for initial IL-33 levels could have been influenced by a few markedly elevated values. This variability, together with the relatively small sample size, might have affected the precision of our statistical estimates. Third, several methodological factors might have influenced the outcomes. For instance, although samples were processed within a 2-hour window, subtle discrepancies in sample collection timing could potentially influence group associations (MODS vs. no MODS). Additionally, slight variations in IL-33 measurements might have occurred during sample collection in the Bio-Plex assay. Fourth, despite the comprehensive consideration of numerous variables in the multivariate analysis, certain factors might still have been overlooked. Fifth, approximately half of the patients had IL-33 levels below the detection limit, which were replaced with 0 for the corresponding day. This conservative approach might have underestimated the true association between IL-33 and MODS.

Despite those constraints, our study is notable as a rare prospective observational investigation that meticulously examined biomarkers to forecast the onset of MODS in trauma patients. We found that initial IL-33 levels were higher in patients who developed MODS and were associated with MODS in a univariate analysis, although that association was not retained after multivariate adjustment. The DNI and ISS were independent predictors of MODS in all patients, and lactate was predictive in those with ISS ≥16. The small sample size and substantial proportion of undetectable IL-33 values should be considered when interpreting these results. Although IL-33 did not emerge as an independent predictor in this study, the observed patterns suggest that it has potential relevance in assessing trauma severity, warranting further validation in larger, well-designed studies.

KEY MESSAGES

• Initial interleukin-33 (IL-33) levels tended to be higher in trauma patients who developed multi-organ distress syndrome (MODS) than in those who did not and then decline rapidly, suggesting that IL-33 could have potential as a biomarker for MODS prediction.

• Although IL-33 levels correlated with injury severity, they were not an independent predictor of MODS; instead, the Injury Severity Score, delta neutrophil index, and lactate levels were identified as significant predictors.

NOTES

CONFLICT OF INTEREST

No potential conflict of interest relevant to this article was reported.

FUNDING

This work was supported by the National Research Foundation of Korea (NRF) under grant number 2022R1I1A1A01068599.

ACKNOWLEDGMENTS

The order of authorship among the co-first authors was determined based on their relative contributions to the research.

AUTHOR CONTRIBUTIONS

Conceptualization: SA, ISS, MJK, DKK, MHR, CSK, KK. Methodology: SA, MJK, MHR, CSK, KK. Software: SA, KK. Validation: SA, ISS, MJK, KK. Formal analysis: MHR, CSK, KK. Investigation: SA, ISS, MJK, DKK, MHR, CSK, KK. Resources: SA, ISS, MJK, KK. Data curation: SA, ISS, MJK, MHR, KK. Visualization: MHR, KK. Supervision: CSK, KK. Project administration: ISS,MJK. Funding acquisition: KK. Writing – original draft: SA, ISS, KK. Writing – review & editing: SA, ISS, MJK, DKK, MHR, CSK, KK. All authors read and agreed to the published version of the manuscript.

SUPPLEMENTARY MATERIALS

Supplementary materials can be found via https://doi.org/10.4266/acc.002500.

Supplementary Fig. 1.

Heat map of interleukin-33 (IL-33) levels for all participants.

acc-002500-Supplementary-Fig-1.pdf

Figure 1.

Patient flowchart. MODS: multi-organ distress syndrome; ISS: Injury Severity Score.

Figure 2.

Dot plot of the interleukin-33 (IL-33) levels in all participants.

Figure 3.

Injured patients stratified according to the absence (IL-33N) or presence (IL-33P) of interleukin-33 (IL-33) at initial presentation. Differences in the (A) Injury Severity Score (ISS), (B) Abbreviated Injury Scale 1 (AIS1) score, (C) AIS2 score, (D) AIS3 score, (E) AIS4 score, (F) AIS5 Score, (G) AIS6 score, (H) white blood cell (WBC) count, (I) neutrophil count, (J) delta neutrophil index (DNI), (K) international normalized ratio (INR), (L) number of packed red blood cell (pRBC) transfusions. NS: not significant. a) P<0.01; b) P<0.05.

Table 1.

Characteristics of patients with and without multi-organ distress syndrome

Variable	All participants			Participants with ISS ≥16
Variable	No MODS (n=67) (%)	MODS (n=20) (%)	P-value	No MODS (n=33) (%)	MODS (n=18) (%)	P-value
Age (yr)	56±17	64±17	0.035	58±17	65±18	0.112
Male sex	56 (83.6)	15 (75.0)	0.511	29 (87.9)	13 (72.2)	0.249
Known history
Hypertension	21 (31.3)	5 (25.0)	0.791	12 (36.4)	5 (27.8)	0.756
Diabetes mellitus	10 (14.9)	2 (10.0)	0.725	4 (12.1)	2 (11.1)	1.000
Cerebrovascular disorder	2 (3.0)	0	1.000	0	0	1.000
Liver disease	2 (3.0)	0	1.000	1 (3.0)	0	1.000
Respiratory disease	2 (3.0)	0	1.000	0	0	1.000
Penetrating injury	5 (7.5)	0	0.585	0	0	1.000
SBP (mm Hg)	122.5±33.8	95.3±32.5	0.001	113.6±38.6	94.9±33.2	0.039
ISS	14.6±8.1	26.0±10.2	<0.001	21.2±5.6	27.8±8.9	0.001
AIS1	0.5±1.1	1.4±1.6	0.014	0.9±1.3	1.4±1.6	0.285
AIS2	0.2±0.6	0.7±1.0	0.083	0.3±0.7	0.7±1.1	0.158
AIS3	1.6±1.5	2.5±1.5	0.023	2.4±1.4	2.7±1.4	0.380
AIS4	1.8±1.3	1.7±1.4	0.783	2.0±1.3	1.7±1.4	0.387
AIS5	1.0±1.4	2.0±1.9	0.037	1.6±1.5	2.2±1.9	0.240
AIS6	0.6±0.7	0.7±0.5	0.523	0.7±0.7	0.7±0.5	0.849
Initial laboratory findings
DNI (%)	1.9±2.0	4.5±3.9	0.003	2.2±2.5	4.2±3.4	0.036
WBC (×10⁹/L)	14.4±5.5	13.9±9.1	0.250	16.1±5.8	14.6±9.3	0.150
Neutrophil (×10⁹/L)	12.1±5.1	11.1±8.5	0.138	13.4±5.3	11.5±8.8	0.073
Creatinine (mg/dl)	1.0±1.0	1.4±0.6	<0.001	1.2±1.4	1.4±0.7	0.014
Hemoglobin (g/dl)	12.4±2.0	11.3±2.3	0.025	12.1±2.1	11.2±2.3	0.058
Platelets (10⁹/L)	222.4±92.5	213.9±81.5	0.762	214.8±115.2	215.9±85.9	0.622
INR	1.1±0.2	1.2±0.2	<0.001	1.1±0.2	1.2±0.2	0.017
CRP (mg/dl)	1.1±2.0	1.9±6.6	0.285	1.0±1.7	0.5±0.3	0.072
Lactate (mmol/L)	2.7±2.0	5.7±4.0	<0.001	3.3±2.5	6.1±4.1	0.003
Procedure			1.000			1.000
Observation only	45 (67.2)	14 (70.0)		27 (81.8)	13 (72.2)
Angioembolization	3 (4.5)	0		2 (6.1)	0
Surgery	19 (28.4)	6 (30.0)		4 (12.1)	5 (27.8)
Worst SOFA score	1.9±1.8	7.8±2.1	<0.001	2.6±1.9	7.9±2.2	<0.001
RBC transfusions ≤24 hr	1.6±2.9	6.3±6.6	<0.001	2.6±3.8	7.0±6.5	0.007
Hospital LOS	18.6±18.5	34.8±27.9	0.016	24.5±22.8	37.1±28.3	0.102
ICU LOS	4.9±8.3	11.7±9.8	<0.001	5.5±5.3	12.4±10.1	0.008
Mortality	0	3 (15.0)	0.011	0	3 (16.7)	0.039

Values are presented as mean±standard deviation or number (%).

MODS: multi-organ distress syndrome; ISS: Injury Severity Score; SBP: systolic blood pressure; AIS: Abbreviated Injury Scale; DNI: delta neutrophil index; WBC: white blood cell; INR: international normalized ratio; CRP: C-reactive protein; SOFA: Sequential Organ Failure Assessment; RBC: red blood cell; LOS: length of stay; ICU: intensive care unit.

Table 2.

Difference in the daily change of IL-33 levels in patients with and without multi-organ distress syndrome

Variable	IL-33 (pg/ml)
Variable	Day 0	Day 1	Day 2	Day 3
All patients
No MODS	4.0±13.8	8.9±30.9	8.5±29.1	7.7±31.1
MODS	12.9±38.6	5.3±6.5	4.7±7.4	2.0±6.0
P-value	0.004	0.132	0.106	0.824
Participants with ISS ≥16
No MODS	3.4±11.3	5.9±18.2	6.1±25.3	5.9±29.1
MODS	13.9±40.6	5.8±6.6	4.8±7.8	2.3±6.3
P-value	0.019	0.114	0.139	0.370

Values are presented as mean±standard deviation.

IL-33: interleukin-33; MODS: multi-organ distress syndrome; ISS: Injury Severity Score.

Table 3.

Multivariate analysis using a logistic regression model to predict multi-organ distress syndrome

	Odds ratio (95% CI)	P-value
All participants
Age	1.04 (1.00–1.08)	0.082
ISS	1.13 (1.05–1.21)	<0.001
DNI (%)	1.35 (1.09–1.67)	0.006
Participants with ISS ≥16
Age	1.04 (0.99–1.09)	0.084
IL-33 (pg/mL)	1.04 (0.97–1.12)	0.246
Lactate (mmol/L)	1.36 (1.09–1.69)	0.006

ISS: Injury Severity Score; DNI: delta neutrophil index; IL-33: interleukin-33.

REFERENCES

1. Hall MJ, DeFrances CJ, Williams SN, Golosinskiy A, Schwartzman A. National hospital discharge survey: 2007 summary. Natl Health Stat Report 2010;(29):1-20,24. PubMed
2. Gentile LF, Cuenca AG, Efron PA, Ang D, Bihorac A, McKinley BA, et al. Persistent inflammation and immunosuppression: a common syndrome and new horizon for surgical intensive care. J Trauma Acute Care Surg 2012;72:1491-501.Article PubMed PMC
3. Vanzant EL, Lopez CM, Ozrazgat-Baslanti T, Ungaro R, Davis R, Cuenca AG, et al. Persistent inflammation, immunosuppression, and catabolism syndrome after severe blunt trauma. J Trauma Acute Care Surg 2014;76:21-9.Article PubMed
4. Hirsiger S, Simmen HP, Werner CM, Wanner GA, Rittirsch D. Danger signals activating the immune response after trauma. Mediators Inflamm 2012;2012:315941.Article PubMed PMC PDF
5. Fields JK, Günther S, Sundberg EJ. Structural basis of IL-1 family cytokine signaling. Front Immunol 2019;10:1412.Article PubMed PMC
6. Liew FY, Girard JP, Turnquist HR. Interleukin-33 in health and disease. Nat Rev Immunol 2016;16:676-89.Article PubMed PDF
7. Kofoed K, Andersen O, Kronborg G, Tvede M, Petersen J, Eugen-Olsen J, et al. Use of plasma C-reactive protein, procalcitonin, neutrophils, macrophage migration inhibitory factor, soluble urokinase-type plasminogen activator receptor, and soluble triggering receptor expressed on myeloid cells-1 in combination to diagnose infections: a prospective study. Crit Care 2007;11:R38.Article PubMed PMC PDF
8. Suarez-de-la-Rica A, Maseda E, Anillo V, Tamayo E, García-Bernedo CA, Ramasco F, et al. Biomarkers (procalcitonin, C reactive protein, and lactate) as predictors of mortality in surgical patients with complicated intra-abdominal infection. Surg Infect (Larchmt) 2015;16:346-51.Article PubMed
9. Marcatti M, Saada J, Okereke I, Wade CE, Bossmann SH, Motamedi M, et al. Quantification of circulating cell free mitochondrial DNA in extracellular vesicles with PicoGreen™ in liquid biopsies: fast assessment of disease/trauma severity. Cells 2021;10:819.Article PubMed PMC
10. Holloway TL, Nicholson SE, Rani M, Cap AP, Schwacha MG. Toll-like receptor responses are suppressed in trauma ICU patients. J Surg Res 2016;206:139-45.Article PubMed PMC
11. Xu J, Guardado J, Hoffman R, Xu H, Namas R, Vodovotz Y, et al. Il33-mediated ILC2 activation and neutrophil IL5 production in the lung response after severe trauma: a reverse translation study from a human cohort to a mouse trauma model. PLoS Med 2017;14:e1002365. Article PubMed PMC
12. Kong T, Park YS, Lee HS, Kim S, Lee JW, You JS, et al. The delta neutrophil index predicts development of multiple organ dysfunction syndrome and 30-day mortality in trauma patients admitted to an intensive care unit: a retrospective analysis. Sci Rep 2018;8:17515.Article PubMed PMC PDF
13. Küchler AM, Pollheimer J, Balogh J, Sponheim J, Manley L, Sorensen DR, et al. Nuclear interleukin-33 is generally expressed in resting endothelium but rapidly lost upon angiogenic or proinflammatory activation. Am J Pathol 2008;173:1229-42.Article PubMed PMC
14. Pichery M, Mirey E, Mercier P, Lefrancais E, Dujardin A, Ortega N, et al. Endogenous IL-33 is highly expressed in mouse epithelial barrier tissues, lymphoid organs, brain, embryos, and inflamed tissues: in situ analysis using a novel Il-33-LacZ gene trap reporter strain. J Immunol 2012;188:3488-95.Article PubMed PDF
15. Krychtiuk KA, Stojkovic S, Lenz M, Brekalo M, Huber K, Wojta J, et al. Predictive value of low interleukin-33 in critically ill patients. Cytokine 2018;103:109-13.Article PubMed
16. Billiar IM, Guardado J, Abdul-Malak O, Vodovotz Y, Billiar TR, Namas RA, et al. Elevations in circulating sST2 levels are associated with in-hospital mortality and adverse clinical outcomes after blunt trauma. J Surg Res 2019;244:23-33.Article PubMed PMC
17. Sundnes O, Ottestad W, Schjalm C, Lundbäck P, la Cour Poulsen L, Mollnes TE, et al. Rapid systemic surge of IL-33 after severe human trauma: a prospective observational study. Mol Med 2021;27:29.Article PMC PDF
18. Shin IS, Gong SC, An S, Kim K. Biomarkers to predict mortality in patients with Fournier's gangrene admitted to the intensive care unit after surgery in South Korea. Acute Crit Care 2023;38:452-9.Article PMC PDF
19. van Beest PA, Brander L, Jansen SP, Rommes JH, Kuiper MA, Spronk PE, et al. Cumulative lactate and hospital mortality in ICU patients. Ann Intensive Care 2013;3:6.Article PubMed PMC
20. Li X, Yang Y, Zhang B, Lin X, Fu X, An Y, et al. Lactate metabolism in human health and disease. Signal Transduct Target Ther 2022;7:305.Article PubMed PMC PDF
21. Park BH, Kang YA, Park MS, Jung WJ, Lee SH, Lee SK, et al. Delta neutrophil index as an early marker of disease severity in critically ill patients with sepsis. BMC Infect Dis 2011;11:299.Article PubMed PMC PDF
22. Bang HJ, Kim K, Shim H, Kim S, Jung PY, Choi YU, et al. Delta neutrophil index for predicting mortality in trauma patients who underwent emergent abdominal surgery: a case controlled study. PLoS One 2020;15:e0230149. Article PubMed PMC
23. Mehraj V, Ponte R, Routy JP. The dynamic role of the IL-33/ST2 axis in chronic viral-infections: alarming and adjuvanting the immune response. EBioMedicine 2016;9:37-44.Article PubMed PMC

Figure & Data

References

Citations

Citations to this article as recorded by

PubReader
ePub Link

Cite

CITE: export Copy Format; Close

Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

Format:

RIS — For EndNote, ProCite, RefWorks, and most other reference management software
BibTeX — For JabRef, BibDesk, and other BibTeX-specific software

Include:

Citation for the content below
Citation and abstract for the content below

Predictive value of elevated interleukin-33 levels for multi-organ dysfunction syndrome in trauma patients in South Korea: a prospective observational study

Acute Crit Care. 2025;40(4):594-604. Published online November 28, 2025

DOI: https://doi.org/10.4266/acc.002500

XML Download

Figure

Predictive value of elevated interleukin-33 levels for multi-organ dysfunction syndrome in trauma patients in South Korea: a prospective observational study

Figure 1. Patient flowchart. MODS: multi-organ distress syndrome; ISS: Injury Severity Score.

Figure 2. Dot plot of the interleukin-33 (IL-33) levels in all participants.

Figure 3. Injured patients stratified according to the absence (IL-33N) or presence (IL-33P) of interleukin-33 (IL-33) at initial presentation. Differences in the (A) Injury Severity Score (ISS), (B) Abbreviated Injury Scale 1 (AIS1) score, (C) AIS2 score, (D) AIS3 score, (E) AIS4 score, (F) AIS5 Score, (G) AIS6 score, (H) white blood cell (WBC) count, (I) neutrophil count, (J) delta neutrophil index (DNI), (K) international normalized ratio (INR), (L) number of packed red blood cell (pRBC) transfusions. NS: not significant. a) P<0.01; b) P<0.05.

Figure 1.

Figure 2.

Figure 3.

Predictive value of elevated interleukin-33 levels for multi-organ dysfunction syndrome in trauma patients in South Korea: a prospective observational study

Variable	All participants			Participants with ISS ≥16
Variable	No MODS (n=67) (%)	MODS (n=20) (%)	P-value	No MODS (n=33) (%)	MODS (n=18) (%)	P-value
Age (yr)	56±17	64±17	0.035	58±17	65±18	0.112
Male sex	56 (83.6)	15 (75.0)	0.511	29 (87.9)	13 (72.2)	0.249
Known history
Hypertension	21 (31.3)	5 (25.0)	0.791	12 (36.4)	5 (27.8)	0.756
Diabetes mellitus	10 (14.9)	2 (10.0)	0.725	4 (12.1)	2 (11.1)	1.000
Cerebrovascular disorder	2 (3.0)	0	1.000	0	0	1.000
Liver disease	2 (3.0)	0	1.000	1 (3.0)	0	1.000
Respiratory disease	2 (3.0)	0	1.000	0	0	1.000
Penetrating injury	5 (7.5)	0	0.585	0	0	1.000
SBP (mm Hg)	122.5±33.8	95.3±32.5	0.001	113.6±38.6	94.9±33.2	0.039
ISS	14.6±8.1	26.0±10.2	<0.001	21.2±5.6	27.8±8.9	0.001
AIS1	0.5±1.1	1.4±1.6	0.014	0.9±1.3	1.4±1.6	0.285
AIS2	0.2±0.6	0.7±1.0	0.083	0.3±0.7	0.7±1.1	0.158
AIS3	1.6±1.5	2.5±1.5	0.023	2.4±1.4	2.7±1.4	0.380
AIS4	1.8±1.3	1.7±1.4	0.783	2.0±1.3	1.7±1.4	0.387
AIS5	1.0±1.4	2.0±1.9	0.037	1.6±1.5	2.2±1.9	0.240
AIS6	0.6±0.7	0.7±0.5	0.523	0.7±0.7	0.7±0.5	0.849
Initial laboratory findings
DNI (%)	1.9±2.0	4.5±3.9	0.003	2.2±2.5	4.2±3.4	0.036
WBC (×10⁹/L)	14.4±5.5	13.9±9.1	0.250	16.1±5.8	14.6±9.3	0.150
Neutrophil (×10⁹/L)	12.1±5.1	11.1±8.5	0.138	13.4±5.3	11.5±8.8	0.073
Creatinine (mg/dl)	1.0±1.0	1.4±0.6	<0.001	1.2±1.4	1.4±0.7	0.014
Hemoglobin (g/dl)	12.4±2.0	11.3±2.3	0.025	12.1±2.1	11.2±2.3	0.058
Platelets (10⁹/L)	222.4±92.5	213.9±81.5	0.762	214.8±115.2	215.9±85.9	0.622
INR	1.1±0.2	1.2±0.2	<0.001	1.1±0.2	1.2±0.2	0.017
CRP (mg/dl)	1.1±2.0	1.9±6.6	0.285	1.0±1.7	0.5±0.3	0.072
Lactate (mmol/L)	2.7±2.0	5.7±4.0	<0.001	3.3±2.5	6.1±4.1	0.003
Procedure			1.000			1.000
Observation only	45 (67.2)	14 (70.0)		27 (81.8)	13 (72.2)
Angioembolization	3 (4.5)	0		2 (6.1)	0
Surgery	19 (28.4)	6 (30.0)		4 (12.1)	5 (27.8)
Worst SOFA score	1.9±1.8	7.8±2.1	<0.001	2.6±1.9	7.9±2.2	<0.001
RBC transfusions ≤24 hr	1.6±2.9	6.3±6.6	<0.001	2.6±3.8	7.0±6.5	0.007
Hospital LOS	18.6±18.5	34.8±27.9	0.016	24.5±22.8	37.1±28.3	0.102
ICU LOS	4.9±8.3	11.7±9.8	<0.001	5.5±5.3	12.4±10.1	0.008
Mortality	0	3 (15.0)	0.011	0	3 (16.7)	0.039

Variable	IL-33 (pg/ml)
Variable	Day 0	Day 1	Day 2	Day 3
All patients
No MODS	4.0±13.8	8.9±30.9	8.5±29.1	7.7±31.1
MODS	12.9±38.6	5.3±6.5	4.7±7.4	2.0±6.0
P-value	0.004	0.132	0.106	0.824
Participants with ISS ≥16
No MODS	3.4±11.3	5.9±18.2	6.1±25.3	5.9±29.1
MODS	13.9±40.6	5.8±6.6	4.8±7.8	2.3±6.3
P-value	0.019	0.114	0.139	0.370

	Odds ratio (95% CI)	P-value
All participants
Age	1.04 (1.00–1.08)	0.082
ISS	1.13 (1.05–1.21)	<0.001
DNI (%)	1.35 (1.09–1.67)	0.006
Participants with ISS ≥16
Age	1.04 (0.99–1.09)	0.084
IL-33 (pg/mL)	1.04 (0.97–1.12)	0.246
Lactate (mmol/L)	1.36 (1.09–1.69)	0.006

Table 1. Characteristics of patients with and without multi-organ distress syndrome

Values are presented as mean±standard deviation or number (%).

Table 2. Difference in the daily change of IL-33 levels in patients with and without multi-organ distress syndrome

Values are presented as mean±standard deviation.

IL-33: interleukin-33; MODS: multi-organ distress syndrome; ISS: Injury Severity Score.

Table 3. Multivariate analysis using a logistic regression model to predict multi-organ distress syndrome

ISS: Injury Severity Score; DNI: delta neutrophil index; IL-33: interleukin-33.