Two hundred and two patients were included in this study. Baseline characteristics are described in Table 1. Patients with T2D had a higher BMI (27 vs 24 kg/m², p = 0.0003), more hypertension (76% vs. 51%, p = 0.0003) and dyslipidemia (57% vs. 31%, p = 0.0001) than patients in the control group. Adipose tissue imaging biomarkers including CATi (124 vs 97 mL/m², p = 0.02), VATi (709 vs 440 mL/m², p < 0.0001) and SATi (513 vs 368 mL/m², p < 0.01) were significantly higher in the T2D group. We found also more severe coronary artery calcification (CAC-DRS = 3) in patients with T2D (24% vs 10%, p = 0.008). The CT lung severity score was not different between groups. Outcomes were similar across groups. At day 21 after admission, 42 patients (21%) had died from COVID-19, 48 (24%) required intensive care and 112 (55%) were admitted to in a conventional care unit (CMU).
Value of clinical, biological and imaging parameters for the prediction of early ICU admission or Death in the overall population.
The distributions of clinical, biological and imaging parameters according to the outcome are summarized in table 2. In the overall study population, male sex (71 vs 56 %, p=0.47), dyslipidemia (54 vs 34%, p=0.003), hyperleukocytosis (6.8 vs 6, p=0.005), elevated CRP (93 vs 60, p=0.03) and increased T-troponin level (23 vs 15, p=0.0007) were associated with early ICU admission or death. Adipose tissue biomarkers were not statistically different between outcome categories when considering the overall population. The lung CT severity score was also not different between outcome categories.
Value of clinical, biological and imaging parameters for the prediction of early ICU admission or Death in T2D and controls.
Univariate and multivariate analysis of the main clinical, biological and radiological findings are detailed in table 3 and 4 respectively.
Table 3 : Univariate analysis of variables for the prediction of ICU or Death in the overall population and subgroups
|
Overall
(n=202)
|
Controls
(n=108)
|
T2D
(n=94)
|
OR
|
95% CI
|
p
|
OR
|
95% CI
|
p
|
OR
|
95% CI
|
p
|
Age, years
|
1.01
|
1.00-1.04
|
0.14
|
1.03
|
1.00-1.05
|
0.04
|
0.99
|
0.96-1.03
|
0.67
|
Male sex
|
1.91
|
1.06-3.45
|
0.03
|
1.78
|
0.83-3.90
|
0.14
|
2.35
|
0.93-6.38
|
0.07
|
BMI, kg/m2
|
1.01
|
0.95-1.07
|
0.78
|
0.97
|
0.89-1.04
|
0.40
|
1.08
|
0.99-1.19
|
0.09
|
Dyslipidemia
|
2.33
|
1.32-4.12
|
0.003
|
1.93
|
0.85-4.48
|
0.12
|
3.75
|
1.57-9.50
|
0.003
|
CATi, mL/m2 (continuous)
|
1.005
|
1.0003-1.01
|
0.03
|
1.00
|
0.99-1.01
|
0.99
|
1.01
|
1.00-1.02
|
0.001
|
CATi>100 mL/m2 (dichotomous)
|
1.52
|
0.87-2.67
|
0.15
|
0.73
|
0.34-1.57
|
0.42
|
4.71
|
1.78-12.49
|
0.002
|
VATi, mL/m2
|
1.00
|
1.00-1.00
|
0.24
|
1.00
|
1.00-1.00
|
0.94
|
1.001
|
1.00-1.002
|
0.046
|
SATi, mL/m2
|
1.00
|
1.00-1.00
|
0.64
|
1.00
|
1.00-1.00
|
0.30
|
1.00
|
1.00-1.00
|
0.61
|
Severe lung lesions (>50%)
|
1.64
|
0.75-3.69
|
0.22
|
1.17
|
0.38-3.81
|
0.78
|
1.21
|
0.43-3.59
|
0.73
|
CAC DRS=3
|
1.30
|
0.62-2.74
|
0.48
|
3.49
|
0.9-14
|
0.06
|
0.83
|
0.32-2.17
|
0.70
|
Leukocytes, x109/L
|
1.08
|
0.99-1.18
|
0.09
|
1.11
|
0.98-1.28
|
0.09
|
1.05
|
0.94-1.19
|
0.38
|
eGFR, mL/min/1.73 m2
|
0.99
|
0.99-1.00
|
0.14
|
0.99
|
0.98-1.00
|
0.23
|
0.99
|
0.98-1.00
|
0.26
|
CRP, mg/L
|
1.005
|
1.001-1.008
|
0.007
|
1.00
|
1.00-1.01
|
0.06
|
1.01
|
1.00-1.01
|
0.045
|
Troponin-T, ng/L
|
1.00
|
1.00-1.01
|
0.10
|
1.01
|
1.00-1.02
|
0.09
|
1.00
|
1.00-1.01
|
0.42
|
Glycemia, mL
|
1.05
|
0.96-1.14
|
0.27
|
1.29
|
0.93-1.82
|
0.12
|
1.07
|
0.96-1.21
|
0.22
|
Abbreviations: BMI: body max index. CATi: cardiac adipose tissue index. VATi: visceral adipose tissue index. SATi: subcutaneous adipose tissue index. CAC-DRS: Coronary Artery Calcium Data and Reporting System.
In the T2D group (n=94), CATi as a continuous variable (p=0.001) or as a dichotomous variable (OR=4.71, p=0.002), dyslipidemia (OR=3.75, p=0.003) and VATi (p=0.046) were related to ICU admission or death in univariate analysis (Table 3). After adjustment for age, sex and BMI in model 1 and for dyslipidemia, CRP, TAC-DRS=3 (severe coronary artery calcifications) in model 2, CATi remained independently and significantly associated with ICU admission or death (p=0.005) (Table 4).
In the control group (n=108), only advanced age was associated with ICU admission or death in univariate analysis (p=0.04) whereas dyslipidemia, biological markers and imaging biomarkers were not associated with ICU admission or death. In the multivariate analysis, dyslipidemia and CATi were not associated with death or ICU admission in controls (Table 4).
Table 4 : Multivariate models for the prediction of death or ICU admission in T2D and control patients
|
Controls
|
T2D
|
OR
|
95% CI
|
p
|
OR
|
95% CI
|
p
|
MODEL 1
|
Age, years
|
1.04
|
1.01-1.07
|
0.02
|
1
|
0.94-1.02
|
0.29
|
Male sex
|
2.4
|
1.01-5.7
|
0.04
|
1.7
|
0.6-4.7
|
0.34
|
BMI, kg/m2
|
1
|
0.93-1.11
|
0.67
|
1.1
|
0.96-1.17
|
0.25
|
CATi, mL/m²
(continuous)
|
1
|
0.99-1.004
|
0.34
|
1.01
|
1.003-1.02
|
0.005
|
MODEL 2
|
CATi, mL/m² (continuous)
|
1
|
0.99-1.01
|
0.6
|
1.01
|
1.003-1.02
|
0.005
|
Dyslipidemia
|
1.7
|
0.7-4.3
|
0.3
|
3.7
|
1.4-10.2
|
0.008
|
CRP, mg/L
|
1
|
0.999-1.001
|
0.09
|
1
|
0.998-1.01
|
0.1
|
CAC DRS=3 or <3
|
2.5
|
0.6-11
|
0.2
|
0.5
|
0.2-1.5
|
0.2
|
MODEL 3
|
CATi >100 mL/m² (dichotomous)
|
0.6
|
0.3-1.4
|
0.3
|
4.6
|
1.6-14
|
0.004
|
Dyslipidemia
|
1.8
|
0.7-4.5
|
0.2
|
3.8
|
1.4-10
|
0.007
|
CRP, mg/L
|
1
|
0.999-1.01
|
0.1
|
1
|
0.998-1.01
|
0.1
|
CAC DRS=3 or <3
|
2.5
|
0.6-12
|
0.2
|
0.5
|
0.2-1.5
|
0.2
|
Abbreviations: CATi: cardiac adipose tissue index. BMI: Body mass index, CRP: C reactive protein, CAC-DRS: Coronary artery Calcium and Data System
Model 1: Age, Male Sex, BMI, CATi. Controls: p = 0.07. Diabetes: p =0.005.
Model 2: Dyslipidemia, CRP,CAC=3, CATi. Controls: p =0.11. Diabetes: p = 0.0002.
Model 3: Dyslipidemia, CRP,CAC=3, CATi>100 mL/m². Controls: p =0.07. Diabetes: p =0.0001.
Comparison of the different adipose tissue depots (cardiac, subcutaneous and visceral abdominal) with respect to outcome
In the T2D group, CATi (140 vs 99 mL/m², p=0.004) was significantly increased for patients admitted in ICU or deceased compared with those admitted in CMU whereas VATi and SATi were not significantly different (fig. 2). In controls, adipose tissue biomarkers (CATi, VATi and SATi) were not statistically different between patients admitted in CMU and those with a worse prognosis (ICU admission or death) (Figure 2).
In the T2D group, AUC of CATi was 0.67 (p=0.004) with 100 mL/m² as an optimal cut-off value reaching a sensitivity of 83% and a specificity of 50% for the prediction of early ICU admission or death (Fig. 3). In multivariate analysis, an increased CATi >100 mL/m2 was strongly related to ICU admission or death (OR =4.6, p=0.004, table 4) even after adjustment for dyslipidemia, CRP and CAC-DRS=3. For VATi and SATi, AUC were 0.62 (p=0.051) and 0.54 (p=0.57) respectively (Fig. 2). In the control group, AUCs for the three adipose tissue depots were low and non significantly related to ICU or death (Figure 3).