BISAP vs Ranson criteria in acute pancreatitis

What BISAP and the Ranson criteria each measure

BISAP and the Ranson criteria both add up simple findings to predict a severe course of acute pancreatitis, but they differ in size, timing and origin. Ranson and colleagues published their 11 prognostic signs in Surgery, Gynecology and Obstetrics in 1974, after studying 43 clinical and laboratory measurements in 100 patients. Five signs are scored at admission and six more over the next 48 hours. The Bedside Index for Severity in Acute Pancreatitis (BISAP) came from Wu and colleagues in Gut in 2008. It was built with classification and regression tree analysis on 17,992 cases from 212 US hospitals, validated on 18,256 more, and has five items, all taken from the first 24 hours. Both give one point per item present. Each has a cut-off of 3 points: 3 or more of the 11 Ranson signs, or a BISAP of 3 or more out of 5, marks a patient as higher risk.

The Ranson criteria for pancreatitis

The Ranson criteria for pancreatitis are 11 signs, each worth one point: five recorded at admission and six during the first 48 hours. At admission, a point is scored for age over 55 years, a white cell count over 16,000/mm³, a blood glucose over 200 mg/dL, a lactate dehydrogenase (LDH) over 350 U/L and an aspartate aminotransferase (AST) over 250 U/L. Within 48 hours, a point is scored for a fall in hematocrit of more than 10%, a rise in blood urea nitrogen (BUN) of more than 5 mg/dL, a serum calcium below 8 mg/dL, an arterial PaO2 below 60 mmHg, a base deficit above 4 mEq/L and an estimated fluid sequestration above 6 litres. Ranson designed the original list mainly around alcohol-related pancreatitis and later published modified thresholds for gallstone pancreatitis, such as age over 70 and a white cell count over 18,000/mm³. In his series, mortality rose as the number of positive signs increased.

For comparison, BISAP gives one point each for a BUN above 25 mg/dL (8.9 mmol/L), impaired mental status, two or more SIRS criteria, age over 60 years and a pleural effusion on imaging, all in the first 24 hours. The BISAP score calculator lists the SIRS criteria and the mortality Wu observed at each score.

24 hours vs 48 hours: when each score is ready

Timing is the main practical difference between BISAP and the Ranson criteria. Every BISAP item comes from the first 24 hours: a routine blood test, the vital signs, the mental state, the age and a chest image. Wu's 2011 review in CMAJ describes the score as straightforward to calculate at any time during those first 24 hours. The Ranson criteria can only be completed at 48 hours, because six of the eleven signs describe what happens over the first two days, such as the fall in hematocrit, the rise in BUN and the fluid sequestration. The same review lists that 48-hour wait as the main drawback of the Ranson criteria and judges them no longer as useful in routine practice. Ranson also needs an arterial blood gas for the PaO2 and the base deficit. BISAP, for its part, needs a chest image to score the pleural effusion point.

Which predicts severity and mortality better?

In head-to-head studies, neither BISAP nor the Ranson criteria clearly wins, and the gap between them is small next to their shared limits. Papachristou and colleagues followed 185 patients prospectively in the US: the area under the ROC curve for severe acute pancreatitis was 0.81 for BISAP and 0.94 for Ranson, and the authors judged BISAP's accuracy similar to the older scores. Khanna and colleagues, in 72 patients in India, found 0.80 for BISAP and 0.85 for Ranson for severe disease, and 0.83 and 0.84 for death. Pooling 12 cohorts, Gao and colleagues found the Ranson criteria more sensitive and BISAP more specific. Capurso and colleagues pooled 18 studies of 6,740 patients that measured both: with a pre-test probability of severe disease of 18.7%, a positive score raised it to 48% for BISAP and 50% for Ranson, and a negative score lowered it to 7% and 6%. They concluded that the common scores perform poorly for decisions about individual patients.

Mounzer and colleagues compared nine scores for persistent organ failure in two US cohorts and found modest accuracy across the board (areas under the curve of 0.57 to 0.84), with the Glasgow score the best classifier at admission. The revised Atlanta classification now grades the final severity by organ failure lasting more than 48 hours, which neither score measures directly.

Worked example: one patient scored both ways

One patient shows how BISAP and the Ranson criteria land on different sides of their cut-offs. A 62-year-old man has alcohol-related acute pancreatitis. In the first 24 hours he is alert, with a temperature of 37.6 °C, a heart rate of 108, a respiratory rate of 24, a white cell count of 17,200/mm³ and a BUN of 22 mg/dL, and his chest X-ray shows no pleural effusion. Three SIRS criteria are met (heart rate, respiratory rate and white cell count), so BISAP gives one point for SIRS and one for age over 60: a score of 2, below the cut-off of 3, with an observed mortality of 1.9–2.1% in Wu 2008. At admission his glucose is 240 mg/dL, LDH 410 U/L and AST 120 U/L, so Ranson already counts 4 signs: age over 55, white cell count, glucose and LDH. By 48 hours his BUN has risen to 29 mg/dL and his calcium is 7.6 mg/dL, two more signs, for a Ranson score of 6.

His hematocrit (42% to 39%), PaO2, base deficit and fluid balance stay within the Ranson limits. The two scores were built on different cohorts and outcomes, so a mismatch like this is a reason to keep watching for organ failure, not to trust one number. Enter his findings in the BISAP score calculator to see the result and the mortality band for each score.