VA-BLED
A clinical risk score for estimating 1-year esophageal variceal bleeding risk.
VA-BLED is being developed as a web-based clinical risk stratification tool. This version is for internal testing of the calculator interface and model implementation.
Not an externally validated tool. This website is intended for exploratory clinical risk stratification and research use. It does not replace clinical judgment, guideline-based care, or individualized patient management. Refer to current clinical practice guidelines for management of liver cirrhosis and complications of portal hypertension.
Calculator
Enter the required values below. The calculator uses VA-BLED when previous variceal bleeding is absent and automatically switches to VA-BLED History when previous variceal bleeding is present.
The estimated risk is scenario-based after prevalence adjustment to the selected assumed 1-year bleeding prevalence. Not to be interpreted until external validation.
Result
Enter valid values and click Calculate risk. The result will show only the model selected by previous bleeding status.
Interpretation
VA-BLED assigns patients to low-, intermediate-, or high-risk zones using prevalence-specific internally validated out-of-fold cutoffs.
The displayed percentage is a scenario-based estimate after recalibration to the selected assumed 1-year bleeding prevalence. Not to be interpreted as observed real-world event probability until external validation.
The calculator is still in development and should not be used as the sole basis for treatment decisions.
Methods and formula
The model uses Child-Pugh score, current alcohol use, ALT normalized to its upper limit of normal, and previous variceal bleeding status for the history-based model.
After a valid calculation, this section shows the exact active formula for the selected prevalence scenario.
Citation
Please cite the web calculator as:
Denk D, et al. VA-BLED: web calculator for 1-year esophageal variceal bleeding risk stratification. Version 0.3. Available at: https://vabled.org/.
Contact
Dr. Dominic Denk, denk@rz.uni-frankfurt.de