The study population included 321 patients with COVID-19. The demographic, clinical characteristics and laboratory biomarkers are shown in Table 1. The median age of the 321 patients was 63.0 (51.0–70.0) years, ranging from 19 years to 95 years and 180 (56.1%) patients were male (Table 1). Of the 321 patients, 142 (44.2%) had 1 or more coexisting comorbidity (Table 1). Chronic heart disease (35[10.9%]), diabetes (14[10.1%]), chronic lung disease (14[4.4%]), hypertension (85[26.5%]), diabetes mellitus (58[18.1%]) were the most common coexisting comorbidity (Table 1). The most common symptoms on admission were fever (289[90%]) and cough (258[80.4%]), followed by sputum production (117[36.4%]) and fatigue (114[35.5%]) (Table 1).
The results of demographic, clinical characteristics and laboratory biomarkers in survivor and non-survivor groups are compared in Table 1. In univariable analysis, age, male, coexisting comorbidity, fever, sputum production, shortness of breath, and confusion were significantly different between survivor and non-survivor group (Table 1). We observed that white blood cell count, neutrophil count, aspartate aminotransferase, total bilirubin, creatinine, and lactate dehydrogenase were significantly increased, while lymphocyte count and platelet count were significantly decreased in non-survivors compared with survivors. We then further calculated the ratio of lactate dehydrogenase to albumin and found that the value of LAR was remarkably increased in non-survivors (Table 1). For exploring the risk factors of in-hospital death, when all statistically significant variables were analyzed according to multivariable logistic regression, only older age (OR, 1.11; 95% CI, 1.05–1.16), WBC count (OR, 1.26; 95% CI, 1.11–1.44), lymphocyte count (OR, 0.78; 95% CI, 0.62–0.99) and LAR (OR, 1.29; 95% CI, 1.18–1.40) were found to be significantly associated with in-hospital death (Table 2).
Table 1
Demographic, clinical characteristics and laboratory biomarkers of patients with COVID-19 pneumonia on admission
Characteristics
|
Total
(n = 321)
|
Survivors
(n = 269)
|
Non-survivors
(n = 52)
|
P value
|
Age(years)
|
63.0(51.0–70.0)
|
61.0(50.0-68.5)
|
69.0(63.25-79.0)
|
< 0.001
|
Sex, n (%)
|
|
|
|
< 0.001
|
Male
|
180(56.1)
|
138(51.3)
|
42(80.8)
|
|
Female
|
141(43.9)
|
131(48.7)
|
10(19.2)
|
|
Comorbidity, n (%)
|
|
|
|
< 0.001
|
Yes
|
142(44.2)
|
106(39.4)
|
36(69.2)
|
|
No
|
179(55.8)
|
163(60.6)
|
16(30.8)
|
|
Types of comorbidity, n (%)
|
|
|
|
|
Chronic heart disease
|
35(10.9)
|
25(9.3)
|
10(19.2)
|
0.035
|
Chronic lung disease
|
14(4.4)
|
6(2.2)
|
8(15.4)
|
< 0.001
|
Chronic kidney disease
|
6(1.9)
|
2(0.7)
|
4(7.7)
|
0.001
|
Chronic liver disease
|
5(1.6)
|
5(1.9)
|
0
|
0.322
|
Hypertension
|
85(26.5)
|
70(26.0)
|
15(28.8)
|
0.673
|
Diabetes mellitus
|
58(18.1)
|
47(17.5)
|
11(21.2)
|
0.528
|
Carcinoma
|
6(1.9)
|
5(1.9)
|
1(1.9)
|
0.975
|
Clinical symptoms
|
|
|
|
|
Fever(temperature ≥ 37.3℃)
|
289(90.0)
|
237(88.1)
|
52(100)
|
0.009
|
Cough
|
258(80.4)
|
219(81.4)
|
39(75.0)
|
0.287
|
Sputum production
|
117(36.4)
|
105(39.0)
|
12(23.1)
|
0.029
|
Shortness of breath
|
91(28.3)
|
56(20.8)
|
35(67.3)
|
< 0.001
|
Chest pain
|
85(26.5)
|
71(26.4)
|
14(26.9)
|
0.937
|
Myalgia
|
84(26.2)
|
75(27.9)
|
9(26.2)
|
0.112
|
Arthralgia
|
46(14.3)
|
43(16.0)
|
3(5.8)
|
0.054
|
Fatigue
|
114(35.5)
|
95(35.3)
|
19(36.5)
|
0.866
|
Confusion
|
11(3.4)
|
0
|
11(21.2)
|
< 0.001
|
Laboratory findings
|
|
|
|
|
White blood cell count, 109/L
|
5.78(4.62–7.72)
|
5.5(4.4–7.2)
|
9.4(5.8–15.3)
|
< 0.001
|
Neutrophil count, 109/L
|
4.14(2.8–5.9)
|
3.8(2.6–5.4)
|
8.1(4.8–12.6)
|
< 0.001
|
Lymphocyte count, 109/L
|
1.08(0.7–1.5)
|
1.2(0.8–1.6)
|
0.6(0.5–0.8)
|
< 0.001
|
Haemoglobin, g/L
|
127.4 ± 17.5
|
127.9 ± 16.0
|
125.1 ± 24.1
|
0.435
|
Platelet count, 109/L
|
212.0(160.0-295.0)
|
223.0(172.0-308.0)
|
162.0(121.0-215.0)
|
< 0.001
|
ALT, U/L
|
24.0(16.0–37.0)
|
24.0(15.5–36.0)
|
27.0(19.0-46.8)
|
0.053
|
AST, U/L
|
28.0(19.5–42.5)
|
26.0(19.0-36.5)
|
44.5(31.0-69.3)
|
< 0.001
|
Total bilirubin, µmol/L
|
9.1(7.1–12.1)
|
8.6(6.8–11.7)
|
10.4(8.3–15.7)
|
< 0.001
|
Creatinine, µmol/L
|
70.0(58.0–86.0)
|
66.0(56.5–81.5)
|
92.5(71.8–149.0)
|
< 0.001
|
Albumin, g/L
|
35.0 ± 5.4
|
36.1 ± 4.8
|
29.2 ± 4.7
|
< 0.001
|
LDH, U/L
|
282.0(211–402.0)
|
255.0(204.5-340.5)
|
530.0(431.0-671.3)
|
< 0.001
|
LAR
|
7.9 (5.7–12.8)
|
7.1(5.4–10.0)
|
18.6(14.0-25.2)
|
< 0.001
|
Data are presented as mean ± SD, median (IQR) and n (%). P values were calculated by χ² test, independent sample t-test, or Mann-Whitney U test. Abbreviation: COVID-19 = coronavirus diseases 2019. ALT = alanine aminotransferase. AST = aspartate aminotransferase. LDH = lactate dehydrogenase. LAR = lactate dehydrogenase to albumin ratio. |
Table 2
Multivariable logistic regression analysis for risk factors associated with in-hospital death
Characteristics
|
Multivariable OR (95% CI)
|
P value
|
Age
|
1.11(1.05–1.16)
|
< 0.001
|
WBC count
|
1.26(1.11–1.44)
|
< 0.001
|
Lymphocyte count
|
0.78(0.62–0.99)
|
0.044
|
LAR
|
1.29(1.18–1.40)
|
< 0.001
|
Abbreviation: LAR = lactate dehydrogenase to albumin ratio. |
Then, all statistically significant variables were taken as candidates for ROC analysis and the optimal cut-off values calculated by the ROC analysis, and the areas under the curve(AUC) was presented in Fig. 1. ROC analysis showed that LAR had a higher AUC (0.917) than age (0.722), WBC count (0.779), lymphocyte count (0.188) to predict in-hospital death (Table 3). The optimal cut-off values were 12.3 for LAR and the highest specificity and sensitivity were 84.0% and 84.6%.
Table 3
Areas under the curve (AUC) of Age, WBC count, Lymphocyte count and LAR
Test Result Variable(s)
|
Area
|
Std. Error†
|
Asymptotic Sig.‡
|
Asymptotic 95% Confifidence Interval
|
Age
|
0.722
|
0.037
|
< 0.001
|
0.650
|
0.794
|
WBC count
|
0.779
|
0.039
|
< 0.001
|
0.703
|
0.855
|
Lymphocyte count*
|
0.812
|
0.034
|
< 0.001
|
0.879
|
0.956
|
LAR
|
0.917
|
0.020
|
< 0.001
|
0.879
|
0.956
|
The test result variables: Age, WBC count, Lymphocyte count, LAR has at least one tie between the positive actual state group and the negative actual state group. Statistics may be biased. † Under the nonparametric assumption. ‡ Null hypothesis: true area = 0.5. Lymphocyte count* means lymphocyte count after negative calculation. Abbreviation: LAR= lactate dehydrogenase to albumin ratio. |
Furthermore, to evaluate whether LAR plays a role in predicting disease severity, we also perform another ROC analysis. First, we divided the clinical classification of OVID-19 into three groups with different severity of illness : mild (n = 184), severe (n = 80) and critical (n = 57) patients. The results showed that LAR had a higher AUC (0.931) to differentiate critical from mild patients and had a sensitivity of 87.7% and a specificity of 82.1% (Fig. 2a). Besides, LAR had an AUC (0.861) to differentiate critical from severe patients and had a sensitivity of 86.0% and a specificity of 73.8%(Fig. 2b) and the role of LAR to distinguish severe from mild patients was the worst(Fig. 2c).