Inflammatory Biomarkers for Acute Pancreatitis Severity Prediction, 2018-2023

1,981 hospitalized acute pancreatitis patients from a retrospective cohort study (January 2018-December 2023) were analyzed by Huicong Ma. The study evaluated 12 inflammatory indices, finding the C-reactive protein-to-calcium ratio (CCR) had the highest predictive accuracy (AUC=0.768) for moderate-to-severe disease.

Use Cases

• Predicting moderate-to-severe acute pancreatitis risk based on 12 inflammatory biomarker ratios.
• Comparing the predictive performance of biomarkers like NLR, PLR, and SII for clinical outcomes.
• Identifying critical thresholds for biomarkers such as the CCR inflection point at 15.
• Building logistic regression models for severity prediction adjusted for 28 covariates.

Strengths

• Includes 1,981 patient records from a 5-year period (2018-2023).
• Evaluates and compares 12 distinct inflammatory indices for predictive performance.
• Model results are fully adjusted for 28 covariates, and AUC values with confidence intervals are reported.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count for the underlying data table is unknown, which may limit suitability assessment.
• Data may reflect temporal and institutional bias inherent to a single retrospective cohort study.

Provenance

Source

Huicong Ma via figshare.

Collection Method

Retrospective cohort study of hospitalized patients.

Time Range

January 2018 to December 2023.

Freshness

Last updated 2026-03-18 07:42:07; freshness should be verified.

Geography

null