Predictive Factors and Scoring System for Positive Bile Cultures in ERCP-Naive Patients with Acute Cholangitis

Aims This study analyzed predictive factors for positive biliary cultures in ERCP-naive patients diagnosed with acute cholangitis. The objective was to identify significant variables and develop a predictive score to estimate the probability of a positive culture supporting clinical decision-making.Acute cholangitis is a severe biliary tract infection, but not all patients present with positive biliary cultures, complicating diagnosis and treatment.

Methods A retrospective analysis was conducted using univariate and multivariate approaches. Variables included age, jaundice, INR, C-reactive protein (CRP), malignant or benign diagnosis, fever/chills, and Tokyo severity grade. Univariate analysis employed t-tests and chi-squared tests to assess associations, while logistic regression was used for multivariate analysis to identify independent predictors.

Results Univariate analysis revealed significant associations between positive cultures and age (p<0.001), as well as a malignant diagnosis (p<0.001), which was inversely related. Other variables, such as jaundice (p=0.139), INR (p=0.188), and CRP (p=0.351), showed no significant association with positive cultures. Fever/chills and Tokyo severity grade were also not significant predictors. Multivariate analysis confirmed that age (p<0.001) was a strong positive predictor, while a malignant diagnosis remained a significant negative predictor (p=0.009). Other variables, such as jaundice, INR, and Tokyo severity grade, were not independently associated with culture positivity. CRP showed a trend toward significance (p=0.086), warranting further investigation. While clinically relevant, fever and chills did not reach statistical significance in either analysis. Predictive Score: Based on the multivariate model, the Positive Biliary Culture Score (PBCS) was developed using the following formula: PBCS=(-10.0)+(0.25×Age) – (1.44×Jaundice)+(0.004×INR)+(0.017×CRP) – (4.66×Malignant Diagnosis)+(2.49×Fever/Chills) – (1.84×Tokyo Severity); Where the variables are coded as follows: age- patient's age in years, jaundice-1 if jaundice is present, 0 otherwise, INR, CRP level, malignant diagnosis-1 if the diagnosis is malignant, 0 otherwise, fever/chills-1 if fever or chills are present, 0 otherwise, Tokyo severity coded as 1 (Grade I), 2 (Grade II), or 3 (Grade III). The score demonstrated good discriminative power, with an AUC of 0.7205. At the optimal cutoff of 0.58, sensitivity was 68.55% and specificity 68.87%. This makes it a reliable tool for stratifying patients based on their risk of positive cultures.

Conclusions The PBCS provides a practical and accessible method for predicting positive biliary cultures in ERCP-naive patients. While primarily a predictive tool, its potential applications include assisting clinicians in identifying higher-risk patients for microbiological testing. The PBCS can guide decision-making in managing acute cholangitis by integrating clinical and laboratory data.

Publication History

Article published online:
27 March 2025