Correlation of hemoglobin levels and mortality in burn patients.
retrospective_cohort · Level III
Where this comes from
- Record sourced from PubMed, PMID 41579447.
- Also identified by DOI 10.1016/j.burns.2025.107840.
- No licence information is recorded for this record.
- Because redistribution is not established, this page shows the abstract only. Follow the links below for the full text.
Abstract
This study aimed to investigate the correlation between hemoglobin (Hb) levels and 1-year mortality in burn patients in intensive care units. We conducted a retrospective cohort study, extracting relevant data from the American Intensive Care Unit (ICU) Medical Information Database IV (MIMIC-IV) for burn patients. This data included complete blood counts, biochemistry and coagulation parameters (both maximum and minimum values during ICU hospitalization), Simplified Acute Physiology Score II (SAPSII), Acute Physiology ScoreIII (APSIII), Oxford Acute Severity of Illness Score (Oasis), Sequential Organ Failure Assessment (SOFA), comorbidities, and outcome. A restricted cubic spline model was utilized to illustrate the relationship between Hb levels and 1-year mortality risk. Patients were stratified into three groups based on gender and the relationship curve (Female: Hb<10.0 g/dL, 10.0 g/dL≤Hb<11.5 g/dL, Hb≥11.5 g/dL; Male: Hb<11.0 g/dL, 11.0 g/dL≤Hb<12.5 g/dL, Hb≥12.5 g/dL). Logistic regression models were employed to assess one-year mortality risk in each group, while sensitivity analysis was conducted via subgroup analysis and propensity score matching (PSM) analysis. A total of 278 burn patients were included in the study, revealing a one-year mortality rate of 12.2 %. Classifying the patients according to their Hb levels in 3 groups, the numerical mortality rates was different (Q1:22.1 %, Q2:8.9 %, Q3:7.0 %). Whether treating Hb as a continuous or categorical variable, both univariate and multivariate logistic regression analyses indicated a gradual decrease in mortality as Hb levels increased. In models adjusted for all confounding factors, the mortality risk for the second and third Hb groups was significantly lower compared to the first group (OR=0.29, 95 % CI=0.10-0.79, p = 0.016; vs. OR=0.36, 95 % CI=0.14-0.93, p = 0.035). Subgroup analysis displayed consistent correlations between reduced Hb levels and increased mortality across various ages, genders, and comorbidities, including chronic pulmonary disease, cerebrovascular disease, liver disease, kidney diseases, and diabetes. After stratifying patients into two groups (Female: Hb<11.0 g/dL and Hb≥11.0 g/dL; Male: Hb<12.0 g/dL and Hb≥12.0 g/dL) and adjusting for potential confounders using Propensity Score Matching (PSM) analysis, as well as applying propensity score adjustment and weighted pairwise analysis (PA), the results remained stable (OR=0.34, CI=0.14-0.83, p = 0.017). Lower Hb levels in burn patients during ICU hospitalization are associated with higher one-year mortality.
Medical subject headings
- Burns
- Hemoglobins
- Anemia