Table 1.
Baseline patient characteristics at the time point of admission and mean visceral and subcutaneous adipose tissue areas.
Fig 1.
Assessment of visceral and subcutaneous adipose tissue distribution.
The visceral and subcutaneous adipose tissue areas were segmented at the center plane of the 3rd lumbar vertebra on axial CT-scans using a semi-automatically segmentation tool (3D slicer) [13]. (A) An exemplary scan of a patient with excessive visceral (green area) and low subcutaneous (yellow area) adipose tissue is depicted. (B) An exemplary scan of a patient with low visceral (green area) and excessive subcutaneous (yellow area) adipose tissue is depicted.
Fig 2.
The ICU patients’ visceral and subcutaneous adipose tissue areas correlate with BMI and differ between the genders.
(A, B) Both VAT and SAT areas are closely associated with the clinical categories of the patients´ body mass index (BMI). (C) Female ICU patients are characterized by a significantly smaller VAT area compared to male patients, whereas the SAT area is increased in female compared to male patients (D). (E) The patients´ age positively correlates with the VAT area but not the area of the SAT (F). Data are expressed as correlation analyses including Spearman´s rho coefficient (A, B, E, F) or means ± SEM (C, D); *p<0.05, p***<0.001.
Fig 3.
Characteristics of visceral and subcutaneous adipose tissue area in ICU patients with various comorbidities.
(A) The VAT area is increased in ICU patients with arterial hypertension. (B) In contrast to the SAT area does not differ between patients with and without arterial hypertension. (C, D) Both the VAT as well as SAT area is significantly increased in patients with diabetes mellitus type 2. Patients with previous diagnosis of coronary artery disease (E, F) or liver cirrhosis (G, H) do not show differences in VAT or SAT areas. Data are expressed as or means ± SEM; *p<0.05, **p<0.01.
Table 2.
Correlations of VAT and SAT area with markers of inflammation, markers of organ dysfunction, metabolic laboratory markers and clinical scores at day 1 of ICU admission.
Fig 4.
Augmented areas of the VAT are associated with increased duration of mechanical ventilation and prevalence of sepsis in critically ill patients.
(A, B) Sepsis patients show an increased area of VAT but not SAT area compared to non-sepsis patients. (C, D) ICU patients requiring mechanical ventilation are characterized by a significantly increased area of the VAT but no difference of the SAT area. (E, F) There is a significant positive correlation between the VAT, but not the SAT and the duration of mechanical ventilation. Data are expressed as means ± SEM (A, B, E, F) or correlation analyses including Spearman´s rho coefficient (C, D); *p<0.05.
Fig 5.
Neither the VAT nor SAT area predicts short-term survival in critically ill patients.
(A, B) Patients who survive the ICU stay have similar VAT as well as SAT areas compared to ICU non-survivors. Both tissue areas are unaltered in patients who did or did not survive beyond 30 (C+D), 60 (E+F) or 90 days (G+H) after ICU admission. Data are expressed as or means ± SEM.
Fig 6.
The VAT area predicts overall survival in critically ill patients.
(A, B) When using the median of both adipose tissue areas as cut-off values, neither the VAT nor the SAT area cut-off value predicts overall survival as Kaplan Meier curve analysis reveals. (C) Using the optimal cut-off value, patients with a VAT area below 241.4 cm2 show an improved overall survival. (D) In contrast to the VAT area, the optimal cut-off value of the SAT area does not differ between overall survivors and non-survivors.