Influence of laboratory-based frailty index and early frailty trajectory on in-ICU death in patients with sepsis: a retrospective study based on MIMIC IV database

Objective Sepsis is a critical disease with a relatively high mortality rate. Combined frailty will increase the symptom burden of patients with sepsis. At present, there is no study exploring the influence of laboratory-based frailty index (FI-lab) on in-ICU mortality in patients with sepsis, nor has there been any study exploring the influence of frailty change trajectory on in-ICU mortality in patients with sepsis. This study aims to explore to assess the relationship between FI-Lab and in-ICU mortality in patients with sepsis and the impact of the early frailty trajectory of sepsis patients on their in-ICU mortality, thereby providing a basis for clinical practice. Methods The data were derived from the MIMIC IV v3.1 database. According to the inclusion and exclusion criteria, a total of 2,277 patients were finally included. The frailty was evaluated by using FI-lab. The latent category growth model (LCGM) was adopted to identify the change categories of the patients' frailty trajectories. The COX proportional hazards regression model was used to compare frailty and its changing trajectory with the in-ICU mortality rate of patients with sepsis. The Kaplan-Meier survival curve was used to compare the survival probabilities of different trajectories, and the Log-rank test was used to compare the differences in survival curves between groups. Subgroup analysis was used to verify the stability of the results. Results Two categories of frailty change trajectories were identified in patients with sepsis, namely the persistent frailty group and the persistent non-frailty group. After adjusting for confounding factors, the level of frailty of patients with sepsis on the 3rd, 5th, and 7th days of ICU admission was associated with ICU mortality ( P  < 0.001), and the changing trajectory of frailty also showed similar results ( P  < 0.001). Subgroup analysis showed similar results ( P  < 0.05). The Kaplan-Meier analysis curve results showed that the persistent non-frailty group had a higher survival probability compared with the persistent frailty group ( P  < 0.001). Conclusions FI-lab is associated with the in-ICU mortality rate of patients with sepsis. Patients with sepsis have different frailty change trajectories during their stay in the ICU. Clinical medical staff should pay attention to patients who remain in a frailty state continuously to improve the survival rate of such patients.