This article is more than 5 years old. Information may no longer be current.

Benefit of acute pancreatitis severity scores for predicting mortality uncertain

The benefit of acute pancreatitis severity scores remains uncertain due to test characteristics and clinical utility, according to findings published in the Annals of Internal Medicine.

Due to this, additional studies with improved methodological rigor are needed along with development of new scoring systems, researchers reported.

“Although several clinical practice guidelines recommend the use of certain scoring systems in management decisions, the clinical utility of these systems remains uncertain,” Meng-Yang Di, MD, PhD, and colleagues wrote. “Our goal was to quantify the incremental benefit of various severity scoring systems for predicting prognosis, as well as the clinical utility and cost-effectiveness of these systems in influencing patient management and outcomes.”

The researchers evaluated clinical prediction tools by systematic review using Ovid MEDLINE and EMBASE as data sources. Their strategy used disease name (acute pancreatitis), severity scores such as Acute Physiology and Chronic Health Evaluation (APACHE) II and Ranson criteria, and terms associated with study design, including a filter for studies of diagnostic test accuracy and keywords for prognostic studies. Di and colleagues used longitudinal studies that evaluated the prognostic value of at least one clinical severity score in acute pancreatitis and completed dual data extraction and quality assessment.

Of the 4,039 citations assessed, 94 studies, which evaluated 18 scores in 53,547 patients, met the inclusion criteria. Although six studies provided data on incremental predictive values, all studies provided prognostic accuracy data and most scores demonstrated low prognostic accuracy.

The researchers found that the median sensitivity and specificity of APACHE II at a threshold of seven were 100% (range, 68% to 100%) and 63% (range, 21% to 96%), while the Ranson criteria at a threshold of two were 90% (range, 0% to 100%) and 67% (range, 14% to 97%). However, researchers discovered limited evidence regarding the incremental predictive value of the scoring systems or their effect on patient outcomes.

“Despite the many decades since prediction models of mortality in [acute pancreatitis] were proposed, evidence establishing their prognostic accuracy is incomplete,” Di and colleagues wrote. “Furthermore, the available data do not provide clear guidance on which models should be used in specific patient populations and how they should direct specific therapy to achieve improved clinical outcomes or enhance the cost-effectiveness of care.”– by Savannah Demko

Disclosure: The researchers report no relevant financial disclosures.