METHODS: The literature search was implemented in four following databases: Web of Science, Scopus, PubMed/Medline, and Google Scholar, thus, determining studies that measured the effects of walnut consumption on adiponectin, leptin, and glycemic biomarkers levels from 2004 up to December 2019.

MATERIALS AND METHODS: Patients diagnosed with EPN between 2013 and 2020 were retrospectively included. Data from 15 centers (70%) were used to develop the scoring system, and data from 7 centers (30%) were used to validate it. Univariable and multivariable logistic regression analyses were performed to identify independent factors related to mortality. Receiver operating characteristic curve analysis was performed to construct the scoring system and calculate the risk of mortality. A standardized regression coefficient was used to quantify the discriminating power of each factor to convert the individual coefficients into points. The area under the curve was used to quantify the scoring system performance. An 8-point scoring system for the mortality risk was created (range, 0-7).

RESULTS: In total, 570 patients were included (400 in the test group and 170 in the validation group). Independent predictors of mortality in the multivariable logistic regression were included in the scoring system: quick Sepsis-related Organ Failure Assessment score ≥2 (2 points), anemia, paranephric gas extension, leukocyte count >22,000/μL, thrombocytopenia, and hyperglycemia (1 point each). The mortality rate was <5% for scores ≤3, 83.3% for scores 6, and 100% for scores 7. The area under the curve was 0.90 (95% confidence interval, 0.84-0.95) for test and 0.91 (95% confidence interval, 0.84-0.97) for the validation group.