Presepsin as a Biomarker for Risk Stratification and Prognosis of Acute Cholangitis:A Single Center Prospective Cohort Study


 Background: Presepsin is currently a promising biomarker for the early diagnosis and prognosis of acute bacterial infection. Acute cholangitis is caused by bacterial infection and has high morbidity and mortality. The study engaged to assess the grading and prognostic value of presepsin in patients with acute cholangitis. Methods: This study enrolled patients with acute cholangitis in the emergency department from May 1, 2019 to December 20, 2020. The patients were evaluated for severity by the 2018 Tokyo Guidelines for Acute Cholangitis. Patients’ baseline features and routine clinical data were collected. On admission, presepsin, procalcitonin (PCT) and systemic inflammatory response syndrome (SIRS) and sequential organ failure assessment (SOFA) scores were determined; and blood cultures were performed. IBM SPSS software (version 22.0) was used. The comparation of values was performed by Pearson chi-square test or Fisher's exact test or Mann-Whitney U test. The area under the receiver operating characteristic curve (AUC), multivariate logistic regression, and correlation analysis were used to analyze this importance of determining the presepsin levels on admission for acute cholangitis. Results: In total, 330 patients, including 109, 101, and 120 patients classified as having mild, moderate, and severe cholangitis, respectively, were examined. The AUCs of presepsin were 0.713 in predicting severe acute cholangitis, better than those of white blood cell (WBC 0.411), C-reactive protein (CRP 0.615), PCT 0.608, and total bilirubin (T-Bil 0.441) (P<0.05). The AUC of presepsin in predicting 28-day mortality was higher than that of WBC, CRP, PCT, and T-Bil. The presepsin level in the positive blood culture group was higher than that in the negative blood culture group (P=0.000). The P values for correlations of presepsin with SIRS and SOFA scores were 0.002 and 0.000, respectively. Conclusions: Presepsin levels on admission were correlated with SIRS and SOFA scores. Presepsin may predict positive blood culture and 28-day mortality in patients with acute cholangitis, and it is superior to WBC, CRP, PCT, and T-Bil in the risk stratification and prognostic assessment of severe cholangitis. Trial registration: Ethical code of this study is 2018-P2-063-01 and acquired on May, 22, 2018.


Introduction
Acute cholangitis is an acute in ammatory reaction caused by a bacterial infection of the biliary tract [1].
It rapidly progresses and can develop into sepsis or septic shock [2]. Acute cholangitis is currently diagnosed based on clinical symptoms and signs, white blood cell (WBC) counts, C-reactive protein (CRP) and total bilirubin (T-Bil) levels, and imaging ndings. The mortality rate of patients with acute severe cholangitis ranges from 2.7-10% [1,3]. Thus, early risk strati cation and prognostic assessments are important in the emergency treatment of acute cholangitis; however, an ideal predictive biomarker remains unknown.
The effectiveness of presepsin as an infectious biomarker to grade the early severity of acute cholangitis remains unclear [4]. Therefore, this prospective cohort study aimed to compare levels of presepsin with traditional biomarkers in patients with acute cholangitis on admission to evaluate the role of presepsin in the early risk strati cation and prognostic assessment of acute cholangitis.

Study design
This clinical study was carried out in Beijing Friendship Hospital, a 2000-bed teaching hospital with approximately 200,000 emergency admissions per year. The Medical Ethics Committee of Beijing Friendship Hospital approved this study (Ethical code: 2018-P2-063-01). All patients or their relatives signed the informed consent form prior to study participation.

Inclusion And Exclusion Criteria>
All patients included in this study met the diagnostic criteria for acute cholangitis according to the 2018 Tokyo Guidelines for Acute Cholangitis (TG18) [5]. The diagnostic criteria were as follows: (1) systemic in ammation (fever/chill or abnormal WBC counts or elevated CRP levels); (2) cholestasis (T-Bil levels ≥ 34.2 µg/mL or liver enzyme levels > 1.5 times of the upper normal limit); (3) supporting imaging ndings (imaging evidence or etiology of biliary dilatation, such as benign or malignant stenosis, stone, and stent). Participants were excluded if they had liver diseases, such as viral or drug hepatitis, chronic liver disease, chronic renal insu ciency, or any other infections on admission.

Patients
Overall, this study enrolled 330 patients (age > 18 years) diagnosed with acute cholangitis from May 1, 2019 to December 20, 2020. All patients were administered intravenous antibiotics (imipenem and cilastatin in sodium for injection 0.5 g every 8 hours; or cefoperazone sodium and sulbactam sodium 3.0 g every 12 hours; or levo oxacin 0.5 g once daily; or tazobactam sodium/piperacillin sodium for injection 4.5 g every 8 hours) as soon as acute cholangitis was diagnosed according to the TG18.

Examinations
On admission, whole blood (30 mL) was collected from the patients' cubital vein, and the systemic in ammatory response syndrome (SIRS) [6] and sequential organ failure assessment (SOFA) scores [7] were also determined immediately. Blood was tested for the following parameters: WBC, CRP, platelet, procalcitonin (PCT), presepsin, T-Bil, total protein (TP), albumin (A), alanine aminotransferase, aspartate aminotransferase, lactate dehydrogenase, myocardial enzyme, serum creatinine, and international normalized ratio. Blood culture was also performed. Furthermore, 1.5 mL of arterial blood was collected for blood gas analysis.

Data Collection
Clinical data, including baseline demographic features and information related to underlying disease, composite complication index (CCI), antibiotic or uid resuscitation before enrollment, laboratory parameters, and imaging, were collected on admission. Data regarding vital signs, mental status, oxygen ow, hepatic function, biomarkers, blood culture, urinary output, vasoactive drugs (dopamine, dobutamine, adrenaline, and norepinephrine), arterial partial pressure of oxygen, and SIRS and SOFA scores were also collected on admission.

Measurement
Presepsin levels were measured using the PATHFAST analyzer (PATHFAST™ presepsin, Mitsubishi Chemical Medience Corporation, Tokyo, Japan), which is a chemiluminescence immunoassay of venous whole blood based on ethylene diamine tetraacetic acid as an anticoagulant [8]. The detection range of presepsin level was 20-200,000 pg/mL. PCT levels were assayed with a BioMerieux Mini VIDAS immunoassay analyzer (Block Scienti c, Bohemia, NY) from the plasma obtained from blood centrifuged for 15 min at 1000 ×g. The detection range of PCT level was 0.05-200 ng/mL. Information regarding other indicators were obtained from the clinical laboratory data.

Severity Of Acute Cholangitis Based On The Tg18
Severe acute cholangitis was diagnosed if any dysfunction was noted in the cardiovascular, nervous, respiratory, or blood systems, kidneys, or liver. Moderate acute cholangitis was de ned if any two of the following conditions were observed: deviant WBC counts, high fever, age > 75 years, hyperbilirubinemia, and hypoalbuminemia. Mild acute cholangitis was diagnosed if the criteria for the above two grades of acute cholangitis was not met.

Statistical Analyses
IBM SPSS version 22.0 software was used for whole statistical analyses. Two-sided P value < 0.05 was considered statistically signi cant, and the signi cant level for pairwise comparisons among three groups was set to 0.017 (P = 0.05/3). Classi ed variables were expressed in counts and percentages and were compared using the Pearson chi-square test or Fisher's exact test. Skewed distribution variables were expressed in medians (25-75th percentiles), and Mann-Whitney U test was used for comparison between groups. Diagnostic sensitivity, speci city, accuracy, critical value, and area under the receiver operating characteristic (ROC) curve (AUC) were determined by performing a ROC curve analyses to verify the role of determining presepsin levels on admission for acute cholangitis. The optimal critical value was calculated using the Youden index according to the ROC curves. The correlation among the main research indicators was determined using Pearson correlation. Multivariate logistic regression analysis was applied for determining the risk factors for 28-day mortality.

Results
Among the 350 patients with acute cholangitis who were screened, 20 didn't agree to take part in this study; nally, 330 or their relatives signed the informed consent form prior to this study. All patients with acute cholangitis on admission were divided into mild, moderate, and severe groups based on the TG18. The numbers of patients in the above three groups were 109 (33.03%), 101 (30.61%), and 120 (36.36%), respectively (Fig. 1).
The features and outcomes of 330 patients with acute cholangitis are shown in Table 1, including 197 men and 133 women. The median age was 74 years. Age, WBC, CRP, PCT, presepsin, T-Bil, D-Bil, CCI, and SIRS and SOFA scores of moderate and severe cholangitis patients were higher than those of mild cholangitis patients, whereas the opposite was true for TP and A. Clinically, patients with severe cholangitis were more likely to be administered carbapenems. Cephalosporins and quinolones were only given for patients with mild acute cholangitis. Eight deaths occurred within 28 days in patients with severe cholangitis. The 28-day mortality rate of patients with acute cholangitis was 2.42%.
Presepsin as a biomarker for28-day mortality in patients with acute cholangitis Eight patients died within 28 days. Presepsin levels in the survival group were lower than those in the death group (P = 0.015), although no signi cant difference was observed in the WBC count, PCT and T-Bil levels in both groups (P > 0.05). As expected, multivariate logistic regression demonstrated an increasing likelihood of 28-day mortality with presepsin (odds ratio [OR], 1; 95% con dence interval [CI], 1.00-1.01; P = 0.001). Although there was no signi cant difference in P values, the AUC of presepsin in predicting 28-day mortality was 0.752, which was higher than that of WBC (0.602), CRP (0.704), PCT (0.637), and T-Bil (0.552). The optimal critical value of presepsin to predict 28-day mortality was 3333 pg/mL (Fig. 2b,   Fig. 3c).
Presepsin as a biomarker of positive blood culture in patients with acute cholangitis WBC, PCT, and presepsin levels were greater in the positive blood culture group than in the negative blood culture group (P < 0.001); however, CRP and T-Bil levels did not show signi cant difference in both groups (P > 0.05). The ROC curves of biomarkers showed that PCT had the best AUC (0.705), but there was no difference observed between the AUC of PCT and presepsin in predicting positive blood culture (P = 0.127). This could indicate a positive blood culture when the presepsin level was > 2222 pg/mL (Fig. 2c,  Fig. 3d).

Discussion
For a long time, acute cholangitis has been diagnosed based on the Charcot's triad; however, its misdiagnosis rate was found to be 30-80% [9][10][11]. With the introduction of WBC, CRP, and T-Bil into the TG07, TG13, and TG18, the diagnosis rate improved, but some cases are still misdiagnosed and early risk strati cation assessments are di cult [12]. PCT has been recommended as a biomarker for predicting severe acute cholangitis; however, PCT levels are increased in patients with trauma [13] and adrenocortical failure [14] and are falsely negative in patients with atypical bacterial infections [15]. Early diagnosis and grading of acute cholangitis are still important clinical problems. Presepsin is a promising biomarker of bacterial infectious diseases discovered in recent years, and its sensitivity and accuracy are better than those of PCT, WBC, and CRP [16][17][18]. Based on the abovementioned ndings, we demonstrated that presepsin had higher sensitivity and speci city in risk strati cation and prognosis of acute cholangitis and had the same positive rate in predicting bacterial culture as PCT.
SOFA score ≥ 2 is the diagnostic criterion for sepsis in the new sepsis-3.0 de nitions, which shows the number and severity of organ dysfunction [19]. In this study, presepsin levels in patients with acute cholangitis were closely related to SIRS and SOFA scores (P < 0.002). On comparing presepsin with PCT, WBC, CRP, and T-Bil, we found that presepsin had the best predictive value for determining the severity of cholangitis (all P < 0.01). This suggests that presepsin levels could determine the severity of acute cholangitis to some extent.
Most cases with acute cholangitis are associated with intestinal bacterial infection and elevated bile acid and bilirubin levels, which play a key role in innate immune function activation [20] and improper cytokine or antigen secretion in mononuclear/kupffer cells [21]. CRP and PCT levels are indirect acute-phase biomarkers of host-pathogen systemic responses, because their syntheses require interleukin-1 and interleukin-6, which are secreted by monocytes/macrophages; however, their synthesis is suppressed in patients with acute cholangitis [21]. Presepsin is an infection-speci c biomarker for innate immune responses that can directly cleave from the monocyte/macrophage-speci c CD14 receptor complex once germs or lipopolysaccharide invade the bloodstream of an organism [22,23]. In patients with acute cholangitis, presepsin may be excessively secreted in the peripheral blood because of monocyte endocytosis, which reduces the abundance of CD14 on the cell membrane [24].
In this study, the overall 28-day mortality rate was 2.42%, which was lower than the reported mortality rate of 2.7-10% for acute severe cholangitis [1,3] and may be related to the aggressive and comprehensive treatment and shorter observation time. Some studies showed that presepsin levels have signi cant prognostic values for mortality risk in patients with sepsis, sepsis-related diseases, or bacterial infections [16,25,26]. Some studies also used WBC counts and T-Bil levels as predictors of mortality in patients with acute cholangitis [27]. In this study, presepsin was more positively correlated with SOFA scores, WBC counts, and T-Bil levels. Multivariate logistic regression demonstrated an increasing likelihood of 28-day mortality with increasing presepsin levels in patients with acute cholangitis, whereas WBC counts and CRP and PCT levels were not associated with 28-day mortality. On comparing the AUCs of presepsin with those of WBC, CRP, PCT, and T-Bil in predicting 28-day mortality, although there was no statistical difference, presepsin was found to have the best AUC. Therefore, presepsin levels on admission could help clinicians determine the early risk of death in patients with cholangitis.
Blood cultures are the gold standard for diagnosing bloodstream infectious diseases; however, blood cultures are often negative, and the results are usually obtained after several days [28]. PCT is a biomarker associated with positive blood cultures [29,30]. As a biomarker of infection, presepsin was superior to PCT in risk strati cation and prognostic assessment of acute cholangitis and was as good as PCT in predicting positive bacterial cultures. Presepsin levels can be obtained several days earlier than blood culture, which facilitates early treatment to some extent. Some limitations should be recognized in this study. Firstly, the study was performed at a single hospital with a relatively small sample size; furthermore, there could be some bias in terms of the enrolled patients, and thus, multicenter and large sample studies need to be conducted in the future. Secondly, the etiology and pathogens that were not statistically analyzed could interfere with the presepsin levels in our patients. Thirdly, the duration of illness and previous treatment effects were not considered when patients were enrolled, which could have affected the observation of biomarkers.

Conclusions
Presepsin levels predict the severity of acute cholangitis, positive blood culture, and 28-day mortality and are also correlated with SIRS and SOFA scores in patients with acute cholangitis. Early and timely determination of presepsin levels is helpful in risk strati cation and prognostic assessment of acute cholangitis. Although the result was exciting, larger study should be performed to verify the role of presepsin in acute cholangitis.

Availability of data and materials
The datasets analyzed during the current study are available from the corresponding author on reasonable request.
Ethics approval and consent to participate    Correlation between white blood cell (WBC), C-reactive protein (CRP), procalcitonin (PCT), presepsin (PSP), total bilirubin (T-Bil) and Score and sequential organ failure assessment (SOFA). PSP had the best positively correlated with SOFA among them. P<0.05 was set to signi cant level.