Photo Credit: Rasi Bhadramani

The following is a summary of “Preoperative prediction of postoperative pancreatic fistula after Pancreaticoduodenectomy: Determination and validation of a cut-off value for the Roberts Score,” published in the April 2025 issue of the American Journal of Surgery by Kaisar et al.

POPF remains one of the most serious and potentially life-threatening complications following pancreaticoduodenectomy. Accurate preoperative risk stratification is critical to optimizing patient outcomes. This study aimed to establish and validate a clinically relevant cut-off value for the Roberts Score, a preoperative, multicenter predictive model specifically designed to estimate the risk of clinically relevant POPF.

A total of 582 patients who underwent pancreaticoduodenectomy were included in this retrospective analysis. The cohort was divided into two groups: an exploratory cohort (n = 466) and a validation cohort (n = 116). In the exploratory cohort, the optimal Roberts Score cut-off value for predicting CR-POPF was determined using the Youden index. The predictive ability of this cut-off was subsequently assessed in the validation cohort through univariate and multivariate logistic regression analyses.

The area under the receiver operating characteristic curve for the Roberts Score in the exploratory cohort was 0.768, indicating good discriminatory power. A cut-off value of 0.268 was identified as optimal for predicting the occurrence of CR-POPF. This cut-off demonstrated significant predictive capability when applied to the validation cohort (p < 0.001). Patients with scores above the 0.268 threshold experienced significantly longer durations until drain removal and extended stays in the intensive care unit, reflecting more complicated postoperative courses.

Furthermore, multivariate logistic regression analysis confirmed that exceeding the Roberts Score cut-off was an independent predictor of CR-POPF (p = 0.038). In addition, the individual scoring variables that comprise the Roberts Score were also independently associated with the risk of developing CR-POPF.

In conclusion, both the Roberts Score cut-off and its component variables serve as independent predictors of clinically relevant POPF after pancreaticoduodenectomy. The use of this validated cut-off may enhance the ability of clinicians to preoperatively identify patients with high risk, guide perioperative management strategies, and potentially improve postoperative outcomes. Future prospective studies are warranted to further explore the integration of the Roberts Score into clinical practice and refine strategies for the prevention and early management of POPF.

Source: americanjournalofsurgery.com/article/S0002-9610(25)00178-3/abstract