Association between iron homeostasis and prognosis in patients with chronic obstructive pulmonary disease: a retrospective analysis from MIMIC-IV database

Abstract

Introduction: Accumulating evidence indicates that inflammatory responses can alter iron-related biomarkers, such as serum iron, ferritin, transferrin, and total iron-binding capacity (TIBC). However, in the context of chronic obstructive pulmonary disease (COPD), characterized by airway inflammation, the relationship between its prognosis and iron-related biomarkers has not been comprehensively assessed.

Methods: Clinical data of 611 COPD patients from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database were retrospectively analyzed. Associations between four iron-related biomarkers-serum iron, ferritin, transferrin, and TIBC-and both long-term and in-hospital mortality in patients with COPD were assessed using the Cox model and the Kaplan-Meier survival analysis. Moreover, receiver operating characteristic curves were used to further evaluate the prognostic predictive ability of these indicators.

Results: The results suggested that higher levels of serum iron and ferritin were significantly associated with poor long-term prognosis in COPD patients. However, higher levels of transferrin and TIBC may reduce the risk of long-term mortality, serving as protective factors. Furthermore, to a certain degree, these four indicators possessed predictive value for both long-term and in-hospital mortality in patients with COPD.

Conclusion: This study underscores the critical connection between iron-related biomarkers and the prognosis of COPD patients, contributing valuable insights for risk stratification and clinical management in this demographic. Future studies, both retrospective and prospective, should investigate the effects of dynamic fluctuations in iron-related biomarkers to enhance the treatment and management of COPD.

Keywords: MIMIC-IV database; chronic obstructive pulmonary disease; in-hospital mortality; iron homeostasis; long-term mortality.

FIGURE 1

The flow chart of the included population.

FIGURE 2

The Kaplan-Meier curves assessing the probability of in-hospital survival associated with different levels of iron (A), ferritin (B), transferrin (C) and TIBC (D), and the Kaplan-Meier curves revealing the probability of long-term survival related to different levels of iron (E), ferritin (F), transferrin (G), and TIBC (H).

FIGURE 3

Forest plots of hazard ratios obtained by the univariate Cox regression analysis revealing significant predictors of in-hospital mortality and long-term mortality. HR, Heart rate; RR, Respiratory rate; MBP, Mean blood pressure; SpO₂, Saturation of peripheral oxygen; TIBC, Total iron binding capacity; WBC, White blood cell; BUN, Blood urea nitrogen; AST, Aspartate aminotransferase; ALT, Alanine aminotransferase; SOFA, Sequential organ failure assessment; MV, Mechanical ventilation.

FIGURE 4

The Restricted Cubic Spline (RCS) plots displaying non-linear correlations between iron (A), ferritin (B), transferrin (C), and total iron binding capacity (D) with in-hospital mortality, while simultaneously illustrating the non-linear associations between iron (E), ferritin (F), transferrin (G), and total iron binding capacity (H) with long-term mortality.

FIGURE 5

The receiver operating characteristic (ROC) curves revealing the ability of iron homeostasis-related indicators to predict 3-day (A), 7-day (B) and 28-day (C) in-hospital mortality, as well as 30-day (D), 60-day (E) and 90-day (F) long-term mortality, the ROC curves revealing the ability of the logistic regression model developed using three iron homeostasis-related indicators such as serum iron, ferritin and transferrin to predict in-hospital (G) and long-term (H) mortality, and the additional ROC curves revealing the ability of the logistic regression model developed using multiple indicators such as serum iron, ferritin, transferrin, age, heart rate, respiratory rate, MBP, SpO2, WBC, neutrophils, calcium, potassium, creatinine, BUN, AST, SOFA and mechanical ventilation to predict in-hospital (I) and long-term (J) mortality.