Abstract
<title>Abstract</title> <p>Background: The clinical course of acute cholangitis can diverge after initial treatment despite similar admission severity. A single inflammatory measurement can therefore incompletely characterize early risk. We examined whether serial blood-count-derived inflammatory indices were associated with hospital mortality in patients with repeated measurements. Methods: This retrospective cohort study included adults with acute cholangitis from Medical Information Mart for Intensive Care IV version 3.1 (MIMIC-IV) and the eICU Collaborative Research Database version 2.0 (eICU). Among patients with repeated differential blood counts, we evaluated trajectories of the systemic inflammation response index (SIRI), neutrophil-to-lymphocyte ratio (NLR), and monocyte-to-lymphocyte ratio (MLR) in relation to hospital mortality. Results: The MIMIC-IV trajectory samples included 426 patients for NLR and 422 each for SIRI and MLR, with 50–51 hospital deaths; the corresponding eICU samples included 143, 143, and 167 patients, with 12–15 deaths. Non-survivors more often had TG18 Grade III disease and higher baseline inflammatory indices. In MIMIC-IV, higher log-baseline SIRI (odds ratio 1.61 per standard deviation, 95% confidence interval 1.15–2.26) and an increasing SIRI slope (odds ratio 1.62 per standard deviation, 95% confidence interval 1.16–2.24) were associated with hospital mortality after TG18 grading. Adding these features increased the apparent MIMIC-IV area under the receiver operating characteristic curve from 0.666 to 0.722; the eICU area under the curve was 0.666. In the strict 72-hour landmark analysis, SIRI slope was associated with subsequent 28-day mortality (hazard ratio 2.20 per standard deviation, 95% confidence interval 1.29–3.75; 25 events). NLR estimates had the same direction of association, whereas MLR estimates were weaker. Conclusions: Among repeatedly tested patients with acute cholangitis, higher baseline SIRI and a rising early SIRI trajectory were associated with hospital mortality. Serial SIRI could complement reassessment after TG18 grading. The findings apply to a selected population enriched for intensive care or intensive monitoring and require prospective validation.</p>