The relationship between inflammatory cytokines and in‐hospital complications of acute pancreatitis

Abstract Objective Acute necrotic collection (ANC), acute peripancreatic fluid collection (APFC), pleural effusion, and ascites are common early complications of acute pancreatitis. This study aimed to investigate the relationship between 12 serum cytokines and the early complications and severity of acute pancreatitis (AP). Methods We retrospectively analyzed the clinical data of 307 patients with AP, and divided them into severe group and mild‐to‐moderate group according to the revised Atlanta classification. Propensity score matching was used to control for confounding factors. Binary logistic regression analysis was used to explore the relationship between cytokine levels and early complications of AP. Results Serum levels of interleukin (IL)‐1β, IL‐5, IL‐6, IL‐8, IL‐10, IL‐17, and tumor necrosis factor‐α were significantly higher in the severe acute pancreatitis (SAP) group than in the non‐SAP group (p < .05). After adjusting for confounding factors, the upper quartiles of IL‐6, IL‐8, and IL‐10 were associated with an increased risk of ANC compared with those in the lowest quartile (IL‐6: quartile 3, odds ratio [OR] = 3.99, 95% confidence interval [CI] = 1.95–8.16; IL‐8: quartile 4, OR = 2.47, 95% CI = 1.27–4.84; IL‐10: quartile 2, OR = 2.22, 95% CI = 1.09–4.56). APFC was associated with high serum levels of IL‐6 (quartile 3, OR = 1.32, 95% CI = 1.02–1.72), pleural effusions were associated with high serum levels of IL‐1β, IL‐6, IL‐8, and IL‐10 (IL‐1β: quartile 4, OR = 2.36, 95% CI = 1.21–4.58; IL‐6: quartile 3, OR = 4.67, 95% CI = 2.27–9.61; IL‐8: quartile 3, OR = 2.95, 95% CI = 1.51–5.79; IL‐10: quartile 4, OR = 3.20, 95% CI = 1.61–6.36), and high serum levels of IL‐6 and IL‐10 were associated with an increased risk of ascites (IL‐6: quartile 3, OR = 3.01, 95% CI = 1.42–6.37; IL‐10: quartile 3, OR = 2.57, 95% CI = 1.23–5.37). Conclusion Serum cytokine levels, including IL‐1β, IL‐6, IL‐8, and IL‐10 may be associated with the occurrence of early complications of AP. In daily clinical practice, IL‐6 may be the most worthwhile cytokine to be detected.


| INTRODUCTION
Acute pancreatitis (AP) is one of the most common gastrointestinal diseases, with a variable course that is often difficult to predict in its early stages. 1 About 80% of patients develop mild to moderately severe disease.However, about 20% of patients develop severe acute pancreatitis (SAP), with serious complications and a mortality rate of 20%-30%. 2 Gallstones, hypertriglyceridemia, and alcohol abuse are the most common causes of AP.In China, the proportion of hypertriglyceridemia (HTG)-related pancreatitis is increasing year by year.Uncommon causes include drugs, endoscopic retrograde cholangiopancreatography (ERCP), hypercalcemia, infections, genetics, autoimmune diseases, and (surgical) trauma. 3o far, SAP remains challenging in contemporary intensive care due to its unpredictable development and high mortality.SAP often accompanies severe complications, and the occurrence of complications is often an important reason for prolonging the course of disease, increasing hospitalization time and mortality. 4The complications of AP can be local or distant, immediate or delayed.Interstitial edematous pancreatitis and necrotizing pancreatitis are the main two subtypes of AP.Interstitial edematous pancreatitis is characterized by inflammation and edema of the pancreatic parenchyma and peripancreatic tissues.Necrotizing pancreatitis occurs when this process progresses to cell death.Acute peripancreatic fluid collection (APFC) is a homogeneous fluid located within or near the pancreas, lacking a clear wall, occurring in the early course of AP, mainly within 4 weeks of onset.However, acute necrotic collection (ANC) occurs in necrotizing pancreatitis, consisting of necrotic material and liquid phase components.APFC and ANC that persist after 4 weeks of onset are respectively called pancreatic pseudocyst (PPC) and walledoff necrosis. 5However, in addition to the common local complications of the pancreas, pulmonary complications are an inevitable topic for AP patients.Pleural effusion is one of the common pulmonary complications in AP patients.The latest data show that the incidence of AP combined with pleural effusion is as high as 34%-54.5%. 6,7The potential mechanisms of pleural effusion accumulation in AP are diverse and can be explained by the following reasons: inflammation-induced changes in capillary permeability, diaphragmatic lymphatic obstruction, pleural-pancreatic fistula formation, and sinus formation between the pleural cavity and PPC. 8 Furthermore, up to 30%-40% of AP patients develop ascites, and a large proportion of patients develop mild to moderate ascites in the initial stage of inflammation, which may be secondary to local inflammation and subsequent transperitoneal and vascular exudation. 9,10In the late stage of AP, ascites are rare, usually due to pancreatic necrosis leading to pancreatic duct rupture.In patients with SAP, if clinically significant intra-abdominal hypertension occurs, early drainage of ascites may be required, and up to 60%-80% of patients with SAP may develop ascites. 11arly complications of AP, as key events of disease prognosis, are crucial for AP patients to be identified and intervened early. 12,13However, to our knowledge, only limited clinical studies have focused on the relationship between some serological indicators (such as C-reactive protein [CRP], neutrophil-to-lymphocyte ratio [NLR], etc.) and early complications of AP, and hardly any studies have systematically evaluated the relationship between cytokine profiles and early complications of AP. 14 The activation of the immune system has been identified as a key trigger and regulator of pancreatitis-induced injury.The uncontrolled activation of inflammatory signals and the intense release of inflammatory cytokines resulted in the severity of the patients. 157][18] In this context, We retrospectively analyzed the data of 12 serum cytokines at admission of 307 patients with AP, and further evaluated their relationship with AP severity and early complications.

| Population
This retrospective study collected information from 1606 patients with AP admitted to the Affiliated Hospital of Xuzhou Medical University from June 2020 to March 2023.The study was conducted in accordance with the principles of the Declaration of Helsinki and was approved by the local ethics committee (Clinical Research Professional Ethics Committee of the First Affiliated Hospital of Xuzhou Medical University, China).Approval number: XYFY2023-KL083-01.4) Chest (nonenhanced) and abdominal (nonenhanced and enhanced) computed tomography examination.Finally, 307 patients were confirmed to be included in this study, while 1299 patients were excluded due to the exclusion criteria (Figure 1).Finally, all patients were divided into two groups: severe group and mild to moderate group.

| Laboratory tests
Laboratory tests included white blood cell (WBC) count, neutrophils, lymphocytes, and serum cytokines, including interleukin (IL)-1β, IL-2, IL-4, IL-5, IL-6, IL-8, IL-10, IL-12p70, IL-17, interferon (IFN)-γ, IFN-α, tumor necrosis factor (TNF)-α.The routine blood parameters of the patients were measured by Sysmex XN 2000 automatic hematology analyzer, the blood biochemical parameters were measured by Cobas 8000 automatic biochemical immunoassay analyzer and its matching reagents, the serum inflammatory factors were measured by Beckman Navios flow cytometer, and the reagent kit was 12 cytokine detection kit (multiplex microsphere flow immunofluorescence method; Qingdao Ruisikai Biotechnology Co., Ltd.); the patients' elbow venous blood was collected and sent for testing immediately within 60 min after collection, and the testing was performed by laboratory doctors.

| Data analysis
Statistical analysis was performed using SPSS 26.0 and R (R Core Team, 2021).R: A language and environment for statistical computing.R Foundation for Statistical Computing.URL https://www.R-project.org/) and RStudio (RStudio Team, 2022).RStudio: Integrated Development Environment for R. RStudio, PBC, URL http://www.rstudio.com/).The normality of continuous data was assessed by the Kolmogorov-Smirnov test.For nonnormally distributed variables, data were presented as median (interquartile range).Group differences of continuous data were tested by the Mann-Whitney U test.For categorical variables, group differences were tested by the χ 2 square test and Fisher's exact test.
The predictive ability of cytokines for AP severity was defined by the area under the curve (AUC) in the receiver operating characteristic (ROC) analysis.The optimal cut-off value was based on the maximum Youden index to discriminate sensitivity and specificity.Cytokines as continuous variables (per 1-SD augment) and ordinal categorical variables (based on quartiles, Q1, Q2, Q3, Q4) were used to determine the relationship between cytokines and the presence and adjusted presence of ANC, APFC, pleural effusion, and ascites by binary logistic regression analysis.Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated.We first performed a crude logistic regression analysis to establish Model 1, to examine the unadjusted association between cytokines and outcomes.Next, to eliminate the potential bias caused by confounding factors, we established two multivariate logistic regression models, Model 2 adjusted for age, gender, and body mass index (BMI); and Model 3 for past medical history, to obtain adjusted ORs and 95% CIs.Finally, we performed a sensitivity analysis by propensity score matching (PSM), PSM used 1:2 matching (using 1:2 nearest neighbor matching method, caliper value of 0.15), and the covariates entered into the PSM model were: age, gender, BMI, and past medical history, to re-evaluate the relationship between cytokines and complications.

| Population characteristics
A total of 307 patients with AP were enrolled, with a median age of 45 years.Among these, 144 (46.7%) were males and 163 (53.1%) were females, 115 (37.5%) were diagnosed with MAP, 145 (47.2%) with MSAP, and 47 (15.3%) with SAP.There was no statistically significant difference in gender and age between the two groups, respectively (p = .925,p = .516).For the etiology of AP, there were 94 cases (30.6%) of cholelithiasis, 125 cases (40.7%) of HTG, 27 cases (8.8%) of alcoholism, and 13 cases (4.2%) of post-ERCP.In addition, there were three cases (1%) of trauma and 45 cases (14.7%) of autoimmune and unknown (idiopathic) nature.The baseline characteristics and etiology of the 307 patients with AP are shown in Table 1.The severe group had higher levels of WBC count, neutrophil count, lymphocyte count, NLR, and CRP than the non-SAP group, and lower blood calcium levels than the mild-to-moderate group (p < .001).The SAP group had a higher probability of developing necrosis, pleural effusion, ascites, organ failure and other related complications, admission to the intensive care unit (ICU), and length of hospital stay than the mild-to-moderate group (p < .001).

| The relationship between cytokine levels and the severity AP
By using nonparametric tests to compare the differences in cytokines between the mild-to-moderate group and the SAP group (Table 2), the serum levels of IL-1β, IL-5, IL-6, IL-8, IL-10, IL-17, and TNF-α in SAP group were significantly higher than those in the mild-to-moderate group (p < .05).Among them, the differences of IL-6, IL-8, and IL-10 between the two groups were very significant (p < .001).In addition, hyperlipidemia was the most prevalent etiology in this study, followed by cholelithiasis.In cholelithiasis, compared with the non-SAP group, the SAP group had significant differences in IL-6 and IFN-γ (p < .05),however, IL-1β, IL-5, IL-8, IL-10, IL-17, TNF-α did not show statistical differences between the two groups.In hyperlipidemia, IL-1β, IL-6, IL-8, and IL-10 still had significant differences between the SAP group and non-SAP group (p < .05),while IL-5, IL-17, TNF-α had no statistical differences between the two groups (Figure 2).

| The relationship between cytokine levels and AP complications
High cytokine levels were closely related to the occurrence of ANC, APFC, pleural effusion, and ascites.In the three models, we compared the variables of the higher  3 and Table S1).
To reduce data bias and confounding variables, patients with complications and those without complications were matched at a 1:2 ratio using PSM analysis (Table S4).The cytokines that had statistical differences before PSM also had significant differences after PSM (Tables S5 and S6).

| DISCUSSION
In recent years, many studies have conducted in-depth exploration of the etiology and pathogenesis of AP.Although the causes of AP are diverse, they all have a common pathological process, namely the destruction of acinar cells and the abnormal activation of pancreatic protease induce the "self-digestion" of the pancreas, which ultimately leads to the occurrence of AP. 19 Damaged acinar cells release damage-associated molecular patterns, which are recognized by pattern recognition receptors (PRRs), and PRRs activate tissue and peripheral blood leukocytes and trigger the release of inflammatory cytokines.Inflammatory mediators further secrete by infiltrating immune-related cells, increase vascular permeability, cause neutrophil extravasation and activation, edema and microvascular disorder, and ultimately lead to hypoxia and tissue damage. 20he results of this study showed that the levels of seven out of 12 cytokines were significantly higher in the SAP group than in the mild or moderate AP group, namely IL-1β, IL-5, IL-6, IL-8, IL-10, IL-17, TNF-α.Some studies have pointed out that among the many cytokines, TNF-α is the earliest and most widely biologically effective cytokine, and also a key factor in causing SIRS and complications.Excessive TNF-α enters the blood circulation, and stimulates the production of cytokines (IL-1β, IL-6, IL-8), causing a "waterfall-like cascade reaction," leading to a cytokine storm.2][23] Sternby et al. 24 performed paired comparisons of samples collected at 0-24 and 25-48 h after AP onset, and the results showed that IL-1β, IL-6, IL-8, and IL-10 changed over time, and there were significant differences between different severity groups after 24 h.Ćeranić et al. 25 monitored the serum concentrations of IL-6, IL-8, and IL-10 at the admission of AP patients, and compared with IL-8 and IL-10, IL-6 had the highest predictive value for SAP at admission, which is consistent with our study.
In this study, we compared the cytokine differences between the ANC group and the non-ANC group, and IL-6, IL-8, and IL-10 showed good predictive value for the  occurrence of ANC.In Kostic et al.'s 26 study, the levels of IL-6, IL-8, and IL-10 were significantly elevated in the necrosis group in the first 3 days of AP, which is similar to our study results.In addition, Schneider et al. 27 studied the kinetics of inflammatory mediator release in necrotizing pancreatitis, and pancreatic injury increased continuously during the observation period, IL-10 showed an early peak, and TNF-α, and IL-1β reached peaks at 6 and 9 h, respectively.In addition, in a prospective study of ICU patients with symptoms 48-72 h, IL-1β levels were found to predict pancreatic necrosis with an accuracy of 88%. 28However, in this study, IL-1β and TNF-α did not show good predictive value for the occurrence of ANC.Unfortunately, we noticed that there was little previous prediction for APFC, and in our study, only IL-6 showed predictive value among the 12 cytokines.Such results may be due to the fact that AP patients with AFPC have a milder local pancreatic inflammatory response.It has been reported that the average time of occurrence of pleural effusion is about 4 days after AP onset. 29In this study, we also evaluated the availability of 12 serum cytokines within 48 h of symptom onset to predict AP patients with pleural effusion, and IL-1β, IL-6, IL-8, and IL-10 all showed good predictive value.Some studies have reported that AP patients with acute lung injury had significantly higher levels of IL-1β, IL-6, IL-8, and TNF-α. 30However, in this study, TNF-α did not show value in predicting AP with pleural effusion.In addition, Vasseur et al. 31 found that IL-1β and IL-10 were elevated in AP patients with pleural effusion.In a mouse lung model of acute necrotizing pancreatitis, IL-6 and IL-10 expression increased. 32At the end of this study, we evaluated the predictive value of 12 cytokines for AP patients with ascites, and IL-6, IL-8, and IL-10 still had good value.However, there have been few studies reported on the relationship between cytokines and AP with ascites.
It is noteworthy that in our study, IL-6 had a significant correlation with both SAP and the four complications.In 1973, IL-6 was identified as a soluble factor secreted by T cells, which is important for B cells to produce antibodies. 33In experimental AP, the peak elevation of IL-6 concentration in the systemic circulation was proven to be an early event before severe pancreatic injury. 34Since its discovery more than 50 years ago, the IL-6 pathway has been confirmed as a key pathway involved in many disease disorders.Since the approval of anti-IL-6 receptor therapy, it has been a decade, and this therapy is now used worldwide for various rheumatic diseases, Castleman's disease, and cytokine release syndrome and other diseases. 35It is well known that IL-6 has a dual role, so there is a wide debate on its homeostatic and pathogenic roles in various immune diseases when considering IL-6 as a therapeutic target. 36Previous studies have shown that in the early stage of SAP, IL-6 can exert anti-inflammatory effects through its classical pathway, and when the level of IL-6 in the body is too high, it can exert inflammatory effects through the transduction pathway, and can expand the local pancreatic tissue damage to the lungs. 37ytokines are important multieffect regulators of immune responses, which have proinflammatory and anti-inflammatory properties, and can produce effective defense responses against invading pathogens.On the other hand, cytokines may interfere with immune responses and enhance inflammation.Blocking the activity of proinflammatory cytokines can promote the survival of sepsis animal models, however, this therapeutic strategy has not improved clinical outcomes. 38he pathological and biological characteristics of AP are tissue damage leading to immune dysregulation, and the severity of the disease is related to the degree of immune response.The imbalance of the body's inflammatory response regulation is an important cause of severe or even death in patients with pancreatitis.Later studies have shown that there are two problems with early blockade therapy: one is that there are many types of inflammatory factors in the body of AP patients, and various inflammatory factors constitute a complex interaction network, which are both balanced and complementary.The treatment method of inhibiting one of the inflammatory factors will cause other network members to increase compensatory, resulting in no therapeutic effect.Secondly, the proinflammatory/antiinflammatory response in the AP process is dynamic, and the inflammatory factors and inflammatory cells will also change under different conditions. 39The complexity of the immune response requires fine-tuned treatment intervention, and balanced treatment of proinflammatory and anti-inflammatory responses may be a hopeful treatment option for AP.In the past, there were relatively few studies on the relationship between serum cytokine levels and common complications of AP.The advantage of this study is that it systematically evaluated the correlation between SAP and AP in-hospital complications and 12 serum cytokine levels.However, some limitations of this study cannot be ignored.The sample size of this study was limited, the number of severe patients was small, and there was potential bias.Further studies and confirmation are needed in large-scale, multicenter prospective studies, and the clinical value of different etiologies of AP patients can be further studied in the future.Secondly, as this study was a retrospective study, the sample collection time was a single time point within 48 h after admission, and longitudinal monitoring at different time points in the future can better understand the dynamic changes of AP.Third, the data of this study was limited, and there was a lack of comparison with other clinical inflammatory indicators.
In conclusion, our study revealed that serum cytokine levels, including IL-1β, L-6, IL-8, and IL-10, may be related to the occurrence of early complications of AP.Among them, IL-6 has a relatively good predictive value.For the current era of monoclonal antibodies and other targeted therapies, clarifying the upregulated cytokines in the course of AP may provide reference for the future direction of targeted therapy research.Cytokines, as a cheap, repeatable, and noninvasive systemic inflammatory marker, monitoring IL-6 levels within 48 h of AP admission may help predict the clinical outcomes of these patients.

( 1 )
First diagnosed with AP. (2) Age of at least 18 and not more than 80 years.(3) Patients with test results of 12 cytokines drawn within 48 h of symptom onset.(
According to the revised Atlanta classification in 2012: (1) Mild acute pancreatitis (MAP), without organ dysfunction and local or systemic complications.(2) Moderately severe acute pancreatitis (MSAP), with transient (≤48 h) organ dysfunction and/or local complications.(3) SAP, with persistent (>48 h) organ dysfunction.The diagnosis of organ dysfunction was based on the modified Marshall scoring system, and any organ score ≥2 points could define the presence of organ dysfunction.

F
I G U R E 1 Flow chart.AP, acute pancreatitis; CT, computed tomography; MAP, mild acute pancreatitis; MSAP, moderately severe acute pancreatitis; SAP, severe acute pancreatitis.
Demographic data and clinical characteristics of 307 AP patients.
Comparison of cytokine levels between the MAP and MSAP group and the SAP group.