Current prognostic models in primary biliary cirrhosis have low precision, partly due to the restricted inclusion criteria of some cohorts used for modelling.
Another reason for low precision in the prognostic models is because of the prolonged natural course of the disease.
Dr Chun-Wing Chan and colleagues hypothesized that better precision could be achieved with a staged model.
The research team used ascites or peripheral edema as a new starting-point for prediction.
|Time to Clinical diagnosis of ascites to death was 3 years with the current model|
|Scandinavian Journal of Gastroenterology|
The researchers based the study on an established database of 289 consecutive patients, followed between 1977 and 1998.
Stepwise Cox regression was used to construct a staged model based on 143 patients.
Of the included patients, 111 first developed ascites and 32 had peripheral edema at entry or during subsequent follow-up.
The model was compared with published models using graphical methods and receiver operating characteristics.
The team found that the mean time from clinical diagnosis of ascites or peripheral edema to death was 3 years.
The researchers noted that the independent prognostic variables included log 10 bilirubin, albumin, age, and history of encephalopathy.
Goodness of fit showed that the survival probabilities predicted by the Ascites Stage Model fitted well with the observed data.
The researchers observed that the Ascites Stage Model was a better predictor of survival than the Mayo long-term model.
In addition, the Ascites Stage Model was found to be a better predictor than the Mayo Repeated Patient Visits Model and the Royal Free Primary Biliary Cirrhosis Prognostic Model.
Dr Chan's team concludes, “The Ascites Stage Model gives a better survival estimate for primary biliary cirrhosis patients once they have developed ascites or peripheral edema compared with the current models”
“The Ascites Stage Model also demonstrates an advantage of staged models in diseases with a prolonged natural history.”