Impact of serum ferritin on the efficacy of hydrocortisone in septic patients: a retrospective cohort study.

BACKGROUND

Sepsis is associated with high morbidity and mortality. Whether serum ferritin can guide hydrocortisone therapy in sepsis remains uncertain.

METHODS

A retrospective cohort was constructed from the Medical Information Mart for Intensive Care IV(MIMIC-IV). Septic patients with ferritin measured within 1 day of onset were included. Serum ferritin levels were compared between 28-day survivors and non-survivors. A receiver operating characteristic (ROC) curve was used to identify the optimal cut-off value of serum ferritin for predicting 28-day mortality, stratifying patients into high- and low-ferritin groups. Variables associated with 28-day mortality were selected by least absolute shrinkage and selection operator (LASSO) regression. Inverse probability of treatment weighting (IPTW) was applied to balance baseline differences between groups. The effect of hydrocortisone on 28-day mortality was then analyzed in the weighted cohort.

RESULTS

Among 718 septic patients, 169 died within 28 days. Median serum ferritin was 476ng/ml (Inter Quartile Range [IQR]:197–1190) in survivors versus 818ng/ml (IQR:284–2200) in non-survivors (P < 0.001). The ROC-derived cut-off value was 526ng/ml. Higher ferritin was associated with increased 28-day mortality (Odds Ratio [OR] = 1.97, 95%CI:1.54–2.52; P < 0.001). Hydrocortisone was not associated with 28-day mortality (OR = 1.14, 95%CI:0.90–1.45; P = 0.276). No interaction between serum ferritin and hydrocortisone was observed (P-interaction = 0.882). In both high- and low-ferritin subgroups, hydrocortisone remained unrelated to 28-day mortality (OR = 1.24, 95%CI:0.90–1.72 and OR = 1.07, 95%CI 0.74–1.56, respectively).

CONCLUSIONS

Elevated serum ferritin is associated with higher 28-day mortality in septic patients. Ferritin status does not modify the efficacy of hydrocortisone in sepsis. CLINICAL TRIAL NUMBER: Not applicable.