Association between white blood cell-to-hemoglobin ratio and in-hospital mortality among critically ill patients with pancreatic cancer: a retrospective cohort study based on MIMIC-IV and eICU-CRD.
- 1School of Medicine, Southeast University, Nanjing, China.
- 2Department of General Surgery, Zhongda Hospital Affiliated to Southeast University, Nanjing, China.
- 3Department of Interventional and Vascular Surgery, Zhongda Hospital Affiliated to Southeast University, Nanjing, China.
Abstract
Critically ill patients with pancreatic cancer represent a high-risk population in whom systemic inflammation, anemia, organ dysfunction, and acute physiological derangement may jointly contribute to poor short-term outcomes. The white blood cell-to-hemoglobin ratio (WHR) is an easily obtainable hematologic index that may reflect both inflammatory burden and reduced oxygen-carrying reserve. Whether WHR adds prognostic information beyond routine clinical assessment in this population remains uncertain. This study aimed to evaluate the association between WHR and in-hospital mortality among critically ill patients with pancreatic cancer. This retrospective cohort study used Medical Information Mart for Intensive Care IV (MIMIC-IV) version 3.1 (2008-2022) and electronic ICU Collaborative Research Database (eICU-CRD) version 2.0 (2014-2015). Adults aged ≥18 years with International Classification of Diseases (ICD)-9-CM/ICD-10-CM-coded pancreatic cancer and a first qualifying intensive care unit (ICU) stay of ≥24 hours were included, patients without white blood cell count or hemoglobin data were excluded. WHR was calculated from the earliest available white blood cell count and hemoglobin measurements within the first 24 hours after hospital admission. The primary outcome was in-hospital mortality. Multivariable logistic regression was used for the binary outcome analysis, and 24-hour landmark cause-specific Cox and Fine-Gray competing-risk models were added as key time-to-event sensitivity analyses. Predictive performance and the incremental value of WHR beyond clinical variables and its individual components were also assessed. A total of 638 patients were included (MIMIC-IV, n=253, eICU-CRD, n=385); mean age was 67.5±11.3 years, 347 (54.4%) were male, and 53 (8.3%) died during hospitalization. Patients with WHR ≥1.7 were older and had higher white blood cell counts, lower hemoglobin concentrations, and higher SOFA scores. In the fully adjusted Model IV, WHR ≥1.7 remained associated with in-hospital mortality [odds ratio (OR) =2.75, 95% confidence interval (CI), 1.39-5.42, P=0.004]. Time-to-event analyses were consistent: the cause-specific hazard ratio was 2.34 (95% CI, 1.30-4.21, P=0.004), and the Fine-Gray subdistribution hazard ratio was 2.38 (95% CI, 1.34-4.24, P=0.003). WHR alone showed modest discrimination [area under the curve (AUC) =0.662], adding WHR to the clinical model increased the AUC from 0.786 to 0.796, but the bootstrap AUC difference was not statistically significant (P=0.49). Elevated WHR was associated with increased in-hospital mortality among critically ill patients with pancreatic cancer. WHR may serve as a simple adjunctive risk marker, although its incremental discrimination beyond clinical variables was modest. Further prospective studies are warranted to validate its added prognostic value and clarify its potential role in clinical risk assessment.
Related
- Association of first-24-hour hypernatremia and hyperkalemia co-occurrence with 30-day mortality in critically ill adults: a retrospective cohort study
- Nonlinear association between cockcroft-gault creatinine clearance and all-cause mortality in critically ill cerebrovascular disease patients: a retrospective cohort study
- Association of Body Mass Index with Early Inflammatory and Coagulation Biomarker Patterns and Intensive Care Unit Mortality in Critically Ill Patients with COVID-19: A Retrospective Cohort Study
- Association of triglyceride-glucose index and triglyceride-glucose-body mass index with all-cause mortality in critically ill patients with aortic disease: A retrospective cohort study.
- Association between thymosin α1 therapy and 28-day mortality risk in critically ill patients with tuberculosis-related sepsis: a retrospective cohort study.
BibTeX
@article{song2026association,
title = {{Association between white blood cell-to-hemoglobin ratio and in-hospital mortality among critically ill patients with pancreatic cancer: a retrospective cohort study based on MIMIC-IV and eICU-CRD.}},
author = {Jialin Song and Hanzhe Zhang and Junsheng Li and Tao Pan and Tao Cheng},
journal = {Journal of gastrointestinal oncology},
year = {2026},
month = sep,
doi = {10.21037/jgo-2026-0639},
url = {https://europepmc.org/article/MED/42996841},
note = {CC BY-NC-ND},
}
RIS
TY - JOUR
TI - Association between white blood cell-to-hemoglobin ratio and in-hospital mortality among critically ill patients with pancreatic cancer: a retrospective cohort study based on MIMIC-IV and eICU-CRD.
AU - Jialin Song
AU - Hanzhe Zhang
AU - Junsheng Li
AU - Tao Pan
AU - Tao Cheng
PY - 2026
DA - 2026/09/30
JO - Journal of gastrointestinal oncology
DO - 10.21037/jgo-2026-0639
UR - https://europepmc.org/article/MED/42996841
AB - Critically ill patients with pancreatic cancer represent a high-risk population in whom systemic inflammation, anemia, organ dysfunction, and acute physiological derangement may jointly contribute to poor short-term outcomes. The white blood cell-to-hemoglobin ratio (WHR) is an easily obtainable hematologic index that may reflect both inflammatory burden and reduced oxygen-carrying reserve. Whether WHR adds prognostic information beyond routine clinical assessment in this population remains uncertain. This study aimed to evaluate the association between WHR and in-hospital mortality among critically ill patients with pancreatic cancer. This retrospective cohort study used Medical Information Mart for Intensive Care IV (MIMIC-IV) version 3.1 (2008-2022) and electronic ICU Collaborative Research Database (eICU-CRD) version 2.0 (2014-2015). Adults aged ≥18 years with International Classification of Diseases (ICD)-9-CM/ICD-10-CM-coded pancreatic cancer and a first qualifying intensive care unit (ICU) stay of ≥24 hours were included, patients without white blood cell count or hemoglobin data were excluded. WHR was calculated from the earliest available white blood cell count and hemoglobin measurements within the first 24 hours after hospital admission. The primary outcome was in-hospital mortality. Multivariable logistic regression was used for the binary outcome analysis, and 24-hour landmark cause-specific Cox and Fine-Gray competing-risk models were added as key time-to-event sensitivity analyses. Predictive performance and the incremental value of WHR beyond clinical variables and its individual components were also assessed. A total of 638 patients were included (MIMIC-IV, n=253, eICU-CRD, n=385); mean age was 67.5±11.3 years, 347 (54.4%) were male, and 53 (8.3%) died during hospitalization. Patients with WHR ≥1.7 were older and had higher white blood cell counts, lower hemoglobin concentrations, and higher SOFA scores. In the fully adjusted Model IV, WHR ≥1.7 remained associated with in-hospital mortality [odds ratio (OR) =2.75, 95% confidence interval (CI), 1.39-5.42, P=0.004]. Time-to-event analyses were consistent: the cause-specific hazard ratio was 2.34 (95% CI, 1.30-4.21, P=0.004), and the Fine-Gray subdistribution hazard ratio was 2.38 (95% CI, 1.34-4.24, P=0.003). WHR alone showed modest discrimination [area under the curve (AUC) =0.662], adding WHR to the clinical model increased the AUC from 0.786 to 0.796, but the bootstrap AUC difference was not statistically significant (P=0.49). Elevated WHR was associated with increased in-hospital mortality among critically ill patients with pancreatic cancer. WHR may serve as a simple adjunctive risk marker, although its incremental discrimination beyond clinical variables was modest. Further prospective studies are warranted to validate its added prognostic value and clarify its potential role in clinical risk assessment.
SN - 2078-6891
SN - 2219-679X
ER -