Comment on “High Preoperative Glasgow Prognostic Score Increases a Risk of Hospital Mortality in Elderly Patients With Perihilar Cholangiocarcinoma”: Integrating Multidimensional Frailty Assessment

We read with great interest the article by Kokumai et al., which demonstrated that a high preoperative Glasgow Prognostic Score (GPS) was independently associated with increased hospital mortality in elderly patients undergoing major hepatectomy for perihilar cholangiocarcinoma (PHCC) [1]. The authors are to be congratulated for addressing a clinically significant gap in the perioperative assessment of elderly patients, a population often underrepresented in surgical oncology studies. The integration of an inflammation-based biomarker such as GPS (derived from C-reactive protein and albumin levels) into surgical decision-making is both rational and timely. However, we would like to highlight several methodological and conceptual considerations, as well as suggest further directions that may enhance patient risk stratification and outcomes in this vulnerable group. First, while GPS is a validated tool for predicting oncologic outcomes and mortality across several malignancies [2], it is not sufficient to comprehensively define frailty. Elderly patients often suffer from age-related physiological decline beyond systemic inflammation. Incorporating standardized frailty metrics such as the Clinical Frailty Scale, G8, or objective sarcopenia indices (e.g., psoas muscle area on cross-sectional imaging) could provide more holistic risk assessment. Recent studies in hepatopancreatobiliary surgery have emphasized that frailty—more than chronological age—is a decisive determinant of surgical outcomes [3]. Second, the use of single-point inflammatory measurements may not accurately reflect a patient's baseline inflammatory status, particularly in the context of biliary obstruction, cholangitis, or sepsis. Serial assessment of CRP and albumin levels, or trajectory-based scoring (delta-GPS), may better capture the dynamic nature of systemic inflammation. A recent prospective study suggested that fluctuations in inflammation over time, rather than static preoperative values, are more predictive of postoperative complications in elderly surgical patients [4]. Third, we note that the study did not specify whether preoperative nutritional or anti-inflammatory interventions were undertaken. Evidence suggests that prehabilitation—including nutritional optimization, exercise therapy, and inflammation control—can modulate GPS and potentially reverse frailty before major hepatobiliary surgery [5]. Incorporating interventional approaches would allow future studies to assess whether high GPS represents a modifiable risk factor or an immutable contraindication to surgery in selected patients. Finally, while GPS was predictive of hospital mortality in elderly patients, it was not prognostic in the younger cohort. This suggests a possible interaction between biological age, inflammatory reserve, and recovery capacity. However, interaction terms or subgroup analyses for GPS × age were not presented. Given the small sample size after propensity score matching (n = 61 per group), the statistical power to detect robust multivariate associations may have been limited. In conclusion, Kokumai et al. offer valuable insights into the prognostic role of GPS in elderly PHCC patients. We propose that future research integrate GPS with multidimensional frailty tools, longitudinal inflammatory profiling, and interventional strategies to further refine surgical candidacy and improve outcomes in this growing patient population. The author declares no conflicts of interest. The author has nothing to report.