The Neutrophil-Lymphocyte Ratio as an Early Predictor of Severity and outcome of Post-Endoscopic Retrograde Cholangiopancreatography Pancreatitis

INTRODUCTION

Acute pancreatitis is a particularly serious adverse event linked to endoscopic retrograde cholangiopancreatography (ERCP). The incidence of post-ERCP pancreatitis (PEP) varies between 4% and 15% [1-3]. Though the incidence of severe form of PEP is low (0.3-0.6%) [4], it has been linked with substantial morbidity, prolonged hospital admission, and, occasionally, death. The PEP mortality rate is increasing. This requires an early and prompt prediction of severity [5]

PEP has been described as the occurrence of the following: acute abdominal pain that is characteristic and lasts for at least twenty-four hours after the ERCP procedure with ≥ 3-fold higher than the upper limit of the normal values of serum amylase and/or lipase levels; and/or abdominal computed tomography (CT) that is consistent with acute pancreatitis. Furthermore, PEP was classified as mild, moderate, or severe based on the revised Atlanta classification [6]. Mild PEP cases (patients with PEP that demands being hospitalized for 2-3 days); moderate PEP cases (PEP patients who require hospitalization for 4-10 days); and severe ones (PEP patients requiring hospital admission for more than 10 days and/or the occurrence of any of these conditions: persistent organ failure, pancreatic necrosis, acute fluid collections, surgical or percutaneous intervention, or death) [7]. Even though the risk factors for PEP occurrence have been extensively researched, few studies have looked into indicators or predictors of PEP severity. Systemic inflammatory response syndrome (SIRS) and serum phosphate have been suggested as prognostic markers for PEP [8,9]. Therefore, a need for novel, simple, and early indicators of severe PEP so that patients can be stratified for vigorous hydration, close monitoring, and particular transferring to the intensive care unit (ICU) [10].

The neutrophil-lymphocyte ratio (NLR) has been shown to predict adverse complications of acute pancreatitis unrelated to ERCP [11-14]. Compared to the total white blood cell (WBC) count, the NLR more accurately depicts the inflammatory and immunological response [11]. In patients with acute pancreatitis, the NLR could predict the severity of the disease, local complications, organ failure, as well as the need for ICU admission [11,13]. Nevertheless, the association between severe PEP and the NLR has not been thoroughly studied [10].

The current study aimed to evaluate the prognostic role of the Neutrophil-Lymphocyte Ratio in Post-ERCP Pancreatitis and to investigate its role in predicting severity, and outcome of the patients and identify the risk factors of PEP.

METHODS

This cohort prospective study was conducted on 90 patients who underwent ERCP and developed pancreatitis. Patients were selected from the inpatient and outpatient clinic of the Tropical Medicine Department at Menoufia University Hospital, and the Hepatogastroenterology Department at National Liver Institute, Menoufia University between February 2023 and February 2024. They were 51 males (56.7%) and 39 females (43.3%); their ages ranged between 20 and 80 years with a mean value of (56.8±13.1) years and they were classified according to revised Atlanta classification [6] into two groups: mild/moderate pancreatitis group included 67 patients and severe pancreatitis group included 23 patients calculated NLR at day 0 & day 1, 2 post ERCP.

Inclusion criteria: Patients subsequently diagnosed with PEP after ERCP whatever the cause (including benign or malignant biliary obstruction).

Exclusion Criteria: Patients who had concurrent acute inflammatory conditions (such as pneumonia, liver abscess, acute cholangitis, and cholecystitis) at admission, patients

All patients were subjected to the following: full history taking (age, sex, DM, HTN,  abdominal pain, weight loss, jaundice, previous pancreatitis, previous ERCP, and previous PEP), full clinical assessment: general examination (BP, temperature, pulse, jaundice, lower limb edema) and local examination(abdominal tenderness, liver, spleen, ascites ). Sample collection and laboratory tests were as follows: Aseptic withdrawal of 6 ml of venous blood after overnight fasting was done from each patient and was divided into three tubes: Tube 1: Two ml of blood was allowed to clot at 37 °C, centrifugated & separated serum used for assaying fasting blood sugar, liver enzymes (ALT & AST) & liver function tests (serum bilirubin, albumin), renal function test, serum calcium, pancreatic enzymes(amylase &lipase) (AU480, Beckman colter, USA), CRP by turbidmetry USA). Tube 2: Two ml of blood was collected on the EDTA tube for assaying CBC, calculating NLR at day 0, 1,2 post-ERCP (Sysmex XN 1000, USA). Tube 3: Two ml of blood was collected on a citrate blue tube for assaying prothrombin time, (Stago, USA). Imaging (abdominal ultrasonography, CT Abdomen (pancreatic protocol), MRCP, and ERCP). Patients’ outcome was assessed regarding hospital stay in day, ICU admission, and mortality.

 Statistics

An IBM personal computer executing the Statistical Package of Social Science (SPSS) version 20 (IBM Corporations, 2011), Armonk, NY, and Epi Info 2000 software was used to gather, tabulate, and statistically analyze the data. The subsequent statistics were used. To investigate the relationship between two qualitative variables, the chi-squared test (χ2) was employed. When comparing two groups with quantitative variables that are not normally distributed, the Mann-Whitney test is a test of significance. When comparing more than two groups with quantitative non-parametric variables, the Kruskal Wallis test (K) is utilized. A test of significance for examining the relationship between non-parametric quantitative variables was the Spearman correlation coefficient. Positive (+) or negative (-) correlations are typically found in the results of the correlation coefficient test (r-test). It is employed to measure how strongly two variables have a linear relationship. For multivariate analysis with a binary dependent variable, binary logistic regression was employed. To assess biomarker performance, receiver operator characteristics (ROC) were created, along with the corresponding sites of maximal accuracy for sensitivity and specificity. The test's accuracy is measured by the area under the ROC curve (AUROC). A test is considered flawless if its area is 1, and useless if its area is 0.5.  [0.90-1 = excellent (A) ,0.80-0.90 = good (B) ,0.70-0.80 = fair, (C) 0.60-0.70 = poor, (D) 0.50-0.60 = fail (F)]

A P-value of (>0.05) was considered not statistically significant. A P-value of (≤0.05) was considered statistically significant. A P-value of (≤0.001) was considered statistically highly significant

Sample size estimation Based on a review of past literature (Lee, S et al.,2021), found that NLR on day 2 post-ERCP has a high discriminatory ability for severe PEP. NLR on day 2 > 8.17 (OR 18.29). The least sample size calculated using Statistics and Sample Size Pro Program Version 6 is 32 participants. The power of the study is 80% and the confidence level is 95%.

RESULTS

Table 1 demonstrates the demographic data among studied patients. 67(74.4%) out of 90 patients with acute pancreatitis exhibit mild to moderate pancreatitis, while 23(25.6%) patients show evidence of severe pancreatitis. There was statistically non-significant difference between the two groups regarding gender, age, history of diabetes mellitus or hypertension.

Regarding the baseline investigation (Pre-ERCP) CBC findings showed that there was a statistically significant difference between the two patients’ groups regarding TLC, being lower in the severe pancreatitis group, while Hb concentration, HCT, platelet count, or NLR didn't differ. Additionally, Statistical analysis revealed that there was a significant difference between the two groups regarding baseline bilirubin, direct bilirubin, and ALT values. Where patients with severe pancreatitis show high values compared to Mild/Moderate cases. However, other baseline laboratory investigations didn't differ as shown in Supplementary Table 1.

According to the indication of ERCP in studied patients, cases with malignant etiology are significantly frequent in patients with severe pancreatitis compared to those with mild to moderate presentation. However, there is a statistically non-significant difference between the two patient groups regarding the purpose of ERCP (Figure 1).

Statistical analysis revealed that previous pancreatitis, previous PEP, and higher BMI are significantly more frequent in severe cases compared to mild to moderate ones. Longer duration of ERCP, difficult cannulation, total Sphincterotomy, papillotomy, and pancreatic duct contrast are frequently associated with severe pancreatitis (Figure 2)

 Table 2 shows Post-ERCP assessment. There was a statistically significant difference between the two patient groups regarding Urea, creatinine, BUN    and CRP day 2 while, TLC, CRP day 1, Glucose day 1&2, Amylase and lipase didn't differ. Also, there was a statistically significant different between the two patient groups regarding hospital stay in days.

Roc curve analysis revealed that, at the cut of point 6.05, NLR at day 1 can predict severe pancreatitis with a sensitivity of 91.3% and specificity of 89.6%.  Furthermore, NLR day 2 at the cut-off point of 5.42 can predict severe pancreatitis with a sensitivity of 91.3% and specificity of 94%. (Table 3 and Figure 3)

Statistical analysis showed that the mean value of NLR day 1, and day 2 are significantly higher in pancreatitis patients requiring ICU admission (9.1±1.96& 11.7±3.74) compared to those who don't (3.71±13.8&3.3±1.9). Additionally, within the severe group, NLR day 2 showed significantly higher values in ICU-admitted patients (11.7±3.74) (Table 4 and Figure 4 A). Regarding the relation between NLR values and mortality in pancreatitis patients, these tables showed that the mean value of NLR day 1and 2 are significantly higher in dead (9.8±2.29& 13.4±3.93) compared to survived patients (4.03±2.39& 3.72±2.40). Within the severe pancreatitis group, NLR days 1 and 2 were significantly higher in dead patients (9.8±2.29& 13.4±3.93) (Table 4 and Figure 4 B). There were significant positive correlations between NLR days 1 and 2 and the duration of hospital stay (0.606 &.0.755). In addition, NLR day 2 shows a significant positive correlation with hospital stay within the severe group (0.439), while NLR day 1 doesn’t (Table 4 and Figure 4 C & D).

BMI ≥30, history of previous pancreatitis, history of previous PEP, NLR day 1(≥6.05), NLR day 2(≥5.42), a malignant indication of ERCP, duration of the procedure, difficult cannulation, and pancreatic duct contrast were significant risk factors for severe pancreatitis (Table 5). Multiple logistic regression analysis revealed that NLR day 1 and day 2 are significant independent risk factors for severe pancreatitis (Table 6).

 Table (1). Demographic data among studied patients

c²:  Chi-square test, SD: Standard deviation, IQ: Inter quartile

 Table (2). Post ERCP Investigation

IQ: Inter quartile, SD: Standard deviation, NLR: neutrophil lymphocytic ratio, CRP: C reactive protein, BUN: blood urea nitrogen

 Table (3).  Roc curve analysis

Cut of point 6.05 day 1

Cut of point 5.42 day 2

 Table (4). Relation between NLR and ICU admission, Mortality, and hospital stay in days

Table (5). Univariate analysis for risk factors for severe pancreatitis

Table (6). Multivariate analysis for risk factors for severe pancreatitis