Acute pancreatitis has a variable course.
Accurate early prediction of severity is essential to direct clinical care.
Current assessment tools are inaccurate, and unable to adapt to new parameters.
None of the current systems uses C-reactive protein.
Modern machine-learning tools can address these issues.
Dr Pearce and colleagues and from England retrospectively assessed 370 patients admitted with acute pancreatitis in a 5-year period.
After exclusions, 265 patients were studied.
First recorded values for physical examination and blood tests, aetiology, severity and complications were recorded.
The investigative team used a kernel logistic regression model was used to remove redundant features.
|This model predicts a severe attack with an area under the receiver-operating curve of 0.8|
The investigators identified the relationships between relevant features and outcome.
Bootstrapping was used to make the best use of data and obtain confidence estimates on the parameters of the model.
A model was used with 8 variables including age, C-reactive protein, respiratory rate, pO2 on air, arterial pH, serum creatinine, and white cell count.
This model predicted a severe attack with an area under the receiver-operating characteristic curve of 0.8.
The team found that the optimum cut-off value for predicting severity gave sensitivity and specificity of 0.87 and 0.7 respectively.
The investigators noted that predictions were better than admission APACHE II scores in the same patients and better than historical admission APACHE II data.
Dr Pearce's team commented, “This system for the first time combines admission values of selected components of APACHE II and C-reactive protein for prediction of severe acute pancreatitis.”
“The score is simple to use, and is more accurate than admission APACHE II alone.”
“It is adaptable and would allow incorporation of new predictive factors.”