Differential Prognostic Associations of Monocyte-Containing and Traditional Systemic Inflammatory Indices with Overall Survival After Resection or Palliative Bypass for Pancreatic Cancer: A Retrospective Cohort Study

Background/Objectives: Systemic inflammatory indices may differ prognostically between resectable and palliative pancreatic cancer. We compared associations of the aggregate index of systemic inflammation (AISI), systemic inflammation response index (SIRI), neutrophil-to-lymphocyte ratio (NLR), systemic immune-inflammation index (SII), and monocyte-to-lymphocyte ratio (MLR) with overall survival after pan-creatic resection or palliative bypass. Methods: This single-center retrospective cohort included 240 patients treated during 2013–2023 (99 resections; 141 palliative bypasses). Indices derived from preoperative blood counts were natural-log transformed, standardized to 1 SD, and entered separately into operative-group-stratified Cox models. Group-specific associations and interactions were estimated. Reconstructed AJCC eighth-edition stage was incorporated only in the resection stratum. Interaction p values were Holm-adjusted. AISI was designated primary and SIRI key secondary. Results: Overall, 215 deaths occurred. Median overall survival was 17.9 months after resection and 5.4 months after palliative bypass. AISI was associated with mortality following bypass (adjusted hazard ratio [HR] per 1-SD increase in ln(AISI), 1.54; 95% confidence interval [CI], 1.28–1.86) but not following resection (HR, 0.97; 95% CI, 0.75–1.25; interaction p = 0.003; Holm-adjusted p = 0.010). SIRI showed corresponding HRs of 1.60 (95% CI, 1.30–1.96) and 0.93 (95% CI, 0.72–1.20), respectively (interaction p = 0.001; Holm-adjusted p = 0.004). MLR also demonstrated heterogeneity, whereas NLR and SII did not. The AISI and SIRI findings remained robust in sensitivity analyses. Conclusions: AISI and SIRI demonstrated operative-context-dependent associations, principally after palliative bypass. Because comparable staging was unavailable in this group, residual confounding by disease extent remains possible. Prospective multicenter validation is required before clinical implementation.