Pronounced effects of the sepsis–obesity paradox in elderly and male individuals without septic shock and the role of immune–inflammatory status: an analysis of MIMIC-IV data

Figures

• 

  Fig. 1. Flow diagram

• 

  Fig. 2. (a) As BMI increases, the proportion of 28-day mortality in sepsis patients first decreases but then increases, reaching its lowest point at BMI quartile 3 and showing an increasing trend at BMI quartile 4. (b) Line chart of 28-day mortality for sepsis patients in the 4 BMI ranges. Q1: quartile 1; Q2: quartile 2; Q3: quartile 3; and Q4: quartile 4

• 

  Fig. 3. Before conducting trend analysis, the collinearity between variables was tested, and the correlation between each variable was less than 0.8, meeting the conditions for logistic regression

• 

  Fig. 4. Subgroup analysis of BMI quartiles 1 and 2 revealed a significant interaction effect between BMI and 28-day mortality in elderly (≥ 65 years) individuals (pinteraction<0.05)

• 

  Fig. 5. (a) BMI had a U-shaped protective effect on 28-day mortality in sepsis patients (p < 0.0001). (b) In those under 65 years of age, BMI had no significant effect on sepsis mortality (p > 0.05). (c) For those over 65 years of age, BMI had a U-shaped relationship with sepsis mortality (p < 0.0001), with a protective BMI ranging from 27.8 ~ 41.7 kg/cm2. (d) In males, BMI had a U-shaped protective effect (p < 0.0001), with a protective BMI ranging from 28.4 ~ 37.7 kg/cm2. (e) In females, BMI had no significant protective effect (p = 0.052)

• 

  Fig. 6. (a ~ f) White blood cell count, monocyte percentage, neutrophil percentage, lymphocyte percentage, NLR, and SII in the population under 65 years of age. (g ~ l) White blood cell count, monocyte percentage, neutrophil percentage, lymphocyte percentage, NLR, and SII in individuals over 65 years old

• 

  Fig. 7. (a ~ f) White blood cell count, monocyte percentage, neutrophil percentage, lymphocyte percentage, NLR, and SII of survivors and non-survivors in the population under 65 years of age. (g ~ l) White blood cell count, monocyte percentage, neutrophil percentage, lymphocyte percentage, NLR, and SII of survivors and non-survivors in the population over 65 years of age

• 

  Fig. 8. (a ~ d) The AUROC for the white blood cell count; monocyte, lymphocyte, and neutrophil percentages; NLR; and SII for the prediction of 28-day mortality in sepsis patients across age and sex subgroups. Notably, the neutrophil and lymphocyte proportions and the NLR and SII were better at predicting sepsis mortality

• 

  Fig. 9. (a, b) In the population under 65 years of age, no significant change in sepsis mortality was observed with increasing BMI (p > 0.05). (c, d) Conversely, in the population over 65 years of age, a significant decrease in sepsis mortality was evident with increasing BMI (p < 0.05), exhibiting a U-shaped trend