The objective of this study was to assess the reporting of studies on new prognostic markers of outcome in acute pancreatitis.Methods:
We used MEDLINE searches complemented with perusal of review articles' references to identify eligible English-language studies. We included studies evaluating nonroutine markers for acute pancreatitis. Eligible outcomes included Atlanta criteria, Japanese criteria for severity, multiple/single organ failure, complications, interventional treatment, hospitalization length, and death. We generated a 47-item checklist on Acute Pancreatitis Prognosis by adapting a previously constructed reporting guidance instrument for prognostic tumor markers (REMARK [Reporting Recommendations for Tumor Marker Prognostic Studies]). The checklist addresses the reporting of essential information in prognostic studies.Results:
The 184 identified eligible studies reported on 196 different prognostic markers. One hundred forty-four studies (78.3%) found at least 1 prognostic marker to be nominally statistically significant. Significant improvements over time were seen in the reporting for 17 items, but major deficiencies were noted even in 2004-2009 studies. Particularly, 12 items were reported in less than 10% of studies overall and even within the most recent studies.Conclusions:
Despite some improvements over time, the reporting of important aspects of prognostic studies in acute pancreatitis remains suboptimal. The proposed REMARK-based checklist may help improve the quality and reporting of research in this field.