Among 239 patients with a positive test for RT-PCR SARS-CoV-2, complete fat measurements and survival data were available in 174 patients. Of these patients, 140 (80%) presented CT features of interstitial pneumonia (COVID-19 pneumonia) and were thus included in the final analysis.
Forty-six patients (46/140; 33%) required ICU admission, 14 of them (10%) died within 30 days, and 5 (4%) died within 180 days from ICU admission. The most important and available characteristics of patients are reported in Table 1. The median BMI was 26.2 Kg/m2 (range 18.4–51.9) Kg/m2. The majority of patients were male (n = 85, 61%) and the median age was 61 (range 17–95) years old. The median value of VAT and SAT were 15 (range 2–38) mm and 16 (2–52) mm, respectively. The median VAT/SAT ratio was 0.80 (0.17–5.75). According to gender, the median VAT/SAT ratio, VAT and SAT were 0.71 (range 0.17–3.29), 14 (range 4–26) mm and 20 (range 2–45) mm in female patients; it was 0.91 (range 0.17–5.75), 15 (range 2–38) mm and 15 (range 2–52) mm in males.
Table 1
Clinical and biochemical features of COVID-19 patients.
Characteristic
|
Value
|
Median age, range (years)
|
61 (17–95)
|
Gender, n (%)
Male
Female
|
85 (61%)
55 (39%)
|
ICU hospitalization, n (%)
|
46 (33%)
|
VAT, mm, median (range)
|
15 (2–38)
|
SAT, mm, median (range)
|
16 (2–52)
|
VAT/SAT, mm, median (range)
|
0.80 (0.17–5.75)
|
BMI, Kg/m2, median (range)
|
26.2 (18.4–51.9)
|
LDH, U/l, median (range)
|
295 (154–888)
|
Lactate, mmol/l, median (range)
|
1.04 (0.53–4.51)
|
Creatinine, mg/dl, median (range)
|
0.84 (0.24–2.34)
|
eGFR, ml/min/1.73 m2 median (range)
|
91 (30–120)
|
Leucocytes, x103/mcl
|
5.6 (2.4–16.5)
|
VAT, visceral adipose tissue; SAT, subcutaneous adipose tissue; ICU, intensive care unit; COPD, chronic obstructive pulmonary disease; CRP, C-reactive protein; LDH, lactate dehydrogenase.
In the overall cohort, fat distribution was similar between COVID-19 patients who died compared to those who did not (VAT: 15.3 ± 8.3 mm versus 14.7 ± 7.0 mm, p = 0.78; SAT: 15.2 ± 8.1 mm versus 17.9 ± 8.8 mm, p = 0.08; VAT/SAT 1.25 ± 1.04 versus 1.07 ± 0.86, p = 0.36; respectively).
The overall survival was not different according to VAT (p = 0.94), SAT (p = 0.32) and VAT/SAT ratio (p = 0.64). When dividing in quartiles, we observed that patients who had the lowest SAT (SAT ≤ 11.25 mm) had a reduced survival compared to patients with a SAT higher than 11.25 mm (77% versus 94% at day 30, p = 0.01; 74% versus 91% at day 180, p = 0.01). By contrast, the overall survival remained unrelated to VAT (p = 0.21) or VAT/SAT ratio (p = 0.45) when looking at quartiles (Fig. 1a-c). These findings were also confirmed among patients admitted to ICU, where we observed that a thinner SAT was associated with lower survival, independently of gender or age (p = 0.02), while VAT or VAT/SAT ratio were not associated with overall survival.
Although it was not associated with survival, VAT/SAT ratio showed a non-linear increased risk of ICU admission, which plateaued out and tended for inversion at VAT/SAT > 1.9, independent of age and sex (p = 0.001). A similar non-linear increased risk of ICU admission was observed for VAT (Fig. 2). Supplementary table 1 summarizes the average relative risk (RR) per each 0.1 increase in VAT/SAT ratio.
Furthermore, we found that fat distribution was not related to in-hospital length of stay among survivors (-0.008 < r < 0.187; p > 0.40) but was associated with clinical and biochemical indices of disease severity. Indeed, VAT correlated inversely with leukocyte count (r=-0.204; p = 0.03) and the positive end-expiratory pressure (PEEP) (r=-0.766; p = 0.004), and directly with serum Interleukin-6 (r = 0.709; p = 0.007). SAT was inversely related with the qSOFA score (r=-0.213; p = 0.043) and circulating lactates (r=-0.369; p = 0.034), confirming an inverse association with disease severity.