3.1 Baseline characteristics
Baseline characteristics are shown in Table 1. The average age was 53 years, and 204 (51.6%) were male. Among these cases, fever on admission (263, 66.6%) was the most common symptom. Cough, shortness of breath, fatigue, and sputum production were present in 257 patients (65.1%), 118 patients (29.9%), 107 patients (27.2%), and 102 patients (25.9%), respectively. Headache (36, 9.1%), nausea or vomiting (36, 9.1%), myalgia or arthralgia (34, 8.6%), sore throat (22, 5.6%), and chill (7, 6.7%) were rare in our study. The most frequent comorbidities were hypertension (102, 25.8%) and diabetes (47, 11.3%). The proportion of coronary heart disease, hepatitis B infection, and chronic obstructive pulmonary disease was 6.4% (25/392), 2.3% (9/392), and 1.5% (6/392), respectively.
According to the low value of laboratory reference values of CD4+T cell count, the 395 COVID-19 patients were divided into two groups: lower CD4+T cell group and higher CD4+T cell group. Patients in the lower group were older (55.0±16.5 vs 51.3±14.8, P=0.033), contained more males (115/195 [59.0%] vs 89/111 [44.5%], P=0.004), and more likely to have shortness of breath (75/195 [38.5%] vs 43/200 [21.5%], P<0.001) and fever on admission (141/195 [72.3%] vs 122/200 [61.0%], P=0.017). And, there was no significant difference in the proportion of comorbidities, including hypertension, diabetes, coronary heart disease, hepatitis of B infection, and chronic obstructive pulmonary disease, between the two groups. Analysis of moderate and severe COVID-19 patients alone also showed the same trend (Supplementary table1, Supplementary table 2).
3.2 Laboratory and Radiographic Findings
Of these 395 COVID-19 patients, median (IQR) values of Hs-CRP (5.0 [2.2-22.9] mg/L) and PCT (0.05 [0.03-0.08] ng/ml) were elevated, while the median (IQR) values of lymphocytes count, CD4+T cell count, CD8+T cell count were within standard ranges (Table1). In moderate patients, only the median (IQR) value of Hs-CRP was elevated. (Supplementary table1.). In severe COVID-19 patients, median (IQR) values of Hs-CRP, PCT, and D-dimer were elevated, while the median (IQR) values of lymphocytes count, CD4+T cell count, CD8+T cell count were decreased. (Supplementary table2). According to lung CT (computed tomography, CT) findings, in all COVID-19 patients, the proportion of ground-glass opacity and local patchy shadowing was 48.7% (170/349) and 38.7% (135/349), respectively.
In terms of laboratory findings, compared with patients in higher CD4+T cell group, patients in the lower CD4+T cell group showed lower median lymphocytes count (0.8 (0.6-1.0) vs 1.5 (1.2-1.8), P<0.001, cells×109/L), CD8+T cell count (168.0 [107.0-250.0] vs 322.0 [244.3-443.5], P<0.001, cells/ul), CD4/CD8 (1.4 [1.1-1.9] vs 1.8 [1.4-2.3], P<0.001), but a higher median hypersensitive C-reactive protein (Hs-CRP) (8.2 [5.0-48.5] vs 4.9 [1.1-7.0], P<0.001, mg/L) and procalcitonin (PCT) (0.06 [0.04-0.11 vs 0.04 [0.02-0.06], P<0.001, ng/ml) (Table1). Analysis of moderate and severe COVID-19 patients alone showed that lymphocytes count and CD8+T cell count were more commonly reduced in severe COVID-19 patients. There was no significant change in the proportion of CD4+T cell lower than the lower limit of normal in moderate and severe COVID-19 patients, but the proportion of CD8+T cells lower than the lower limit of normal in moderate and severe COVID-19 patients accounted for 36.0% (71/197) and 51.5% (102/198), respectively. (Figure 2A). The analysis also found that it is the CD8+T cell count that reflects the severity of the patient’s condition, not the CD4+T cell count. (Figure 2B).
In terms of computed tomography findings, in moderate COVID-19 patients, compared with patients in the higher group, patients in the lower group more often represented as local patchy shadowing (45 [47.4%] vs 33 [32.4%], P=0.031). Ground-glass opacity and local patchy shadowing did not differ between the two groups in the entire patient population. (Table1).
3.3 Treatment and Clinical outcome
In all cases, the proportion of use of oxygen inhalation, and mechanical ventilation were 84.3% (328/389), and 7.7% (30/388), respectively. The most common therapy is treatment with antiviral treatment (388/395, 98.2%), followed by antibiotic treatment (179/395, 45.3%), glucocorticoids treatment (94/395, 23.8%), intravenous immunoglobulin treatment (71/395, 18.2%), and only four patients (4/395, 1.0%) were treated with antifungal drugs. During follow-up, 27 patients died (27/395, 6.8%), and the rest were discharged (368/395, 93.2%).
Compared with patients in the higher CD4+T cell group, patients in the lower group needed more oxygen inhalation (174/193 ,90.2% vs 154/196, 78.6%, P=0.002), mechanical ventilation (26/193, 13.5% vs 4/195, 2.1%, P<0.001), antibiotic treatment (112/195, 57.4% vs 67/200, 33.5%, P<0.001) and glucocorticoids treatment (64/195, 32.8% vs 30/200, 15.0%, P<0.001). Other treatments were similar between the two groups, such as antiviral treatment, intravenous immunoglobulin treatment, and antifungal treatment. The case in-hospital death rate was significantly higher in patients with lower CD4+T cell levels than in those with higher CD4+T cell levels (25/195, 12.8% vs 2/200, 1.0%, P<0.001). The detailed treatment of moderate and severe COVID-19 patients was shown in supplementary table1 and supplementary table2.
3.4 Survival curves of in-hospital death
Kaplan-Meier survival curves of the COVID-19 patients grouped by CD4+T cell count are shown in Figure3. The low CD4+T cell group had a higher in-hospital death rate than the high CD4+T cell group during the follow-up period (log rank<0.001). The same trend was also found in severe COVID-19 patients (log rank<0.001). Kaplan-Meier survival analysis was not performed on moderate COVID-19 patients because no patients died during follow-up.
3.5 Results of Cox proportional hazards analyses of in-hospital death
Cox proportional hazard regression analysis was performed to test the associations between the lower CD4+T cell group and in-hospital death for COVID-19 patients (Supplementary Table3). Results of univariate analyses indicated that patients with lower CD4+T cell count exhibited a 13.659-fold increase in in-hospital death compared to patients with higher CD4+T cell count (hazard ratio (HR):13.659; 95% confidence intervals (CI):3.235-57.671). Meanwhile, age, history of hypertension, history of COPD, white blood cell count, lymphocyte count, CD8+T cell lower group (HR: 10.883; 95%CI: 3.277-36.145), required mechanical ventilation or glucocorticoids or intravenous immunoglobulin treatment or antibiotic treatment or antifungal treatment were correlated with the risk of in-hospital death in patients with COVID-19.
Multivariate survival analysis was performed with Cox’s proportional hazard regression model to identify the independent factors correlated with prognosis (Table2). After adjusting for age, sex, and temperature (Mode 1), the HR of the lower CD4+T cell group for in-hospital death was 14.182 (95%CI: 1.884-106.786, P=0.010). After adjusting for a history of hypertension, a history of diabetes, and shortness of breath (Mode 2), the HR of the lower CD4+T cell group for in-hospital death was 13.631 (95%CI: 3.190-58.243, P<0.001). After adjusting for white blood cells, platelet, and creatinine (Mode 3), the HR of the lower CD4+T cell group for in-hospital death was 8.170 (95%CI: 1.877-35.566, P=0.005). After adjusting for hypersensitive C-reactive protein, procalcitonin, and D-dimer (Mode4), the HR of the lower CD4+T cell group for in-hospital death was 10.644 (95%CI: 2.439-46.458, P=0.002). After adjusting for CD8+T cell lower group and lymphocytes count lower group (Mode 5), the HR of the lower CD4+T cell group for in-hospital death was 13.650 (95%CI: 1.976-94.279, P=0.008); Besides, in this model, the HR of the lower CD8+T cell group for in-hospital death was 2.873 (95%CI: 0.771-10.709, P=0.116) after adjusting for other factors, we thus concluded that reduced CD4+T cell was a better predictor of in-hospital death. After adjusting for age, a history of hypertension, shortness of breath, white blood cell count platelet count, D-dimer, and CD4/CD8 (Mode 6), the HR of the lower CD4+T cell count group for in-hospital death was 7.656 (95%CI: 1.610-36.396, P=0.010). Multivariate analysis demonstrated that presenting with lower CD4+T cell count was an independent risk factor for in-hospital death. Variables like age, white blood cell count, and shortness of breath also showed significance for independently predicting in-hospital death in this study (Figure4). Similarly, Cox proportional hazards analyses were also performed on severe COVID-19 patients, and the results also suggested that lower CD4+T cell count was an independent risk factor for in-hospital death (Supplementary table4, Supplementary table5, Supplementary figure1).