Epidemiological and Clinical Features
A total of 166 patients over 70 yo with COVID-19 were hospitalized in our hospital from January 20 to February 15, 2020. Nineteen cases were excluded due to a lack of data on critical information or follow-up. In total, 147 laboratory-confirmed cases of SARS-CoV-2 infection were included in this study. The median age was 76 (IQR 72–81, ranging from 70 to 95) years. Among them, 85 patients (57.8%) were male. The median follow-up time was 25 (IQR 10.5–37) days. The demographic features showed that only 1 case (0.7%) was mild type, 88 cases (59.8%) were moderate type, 42 cases (28.6%) were severe type, and 16 cases (10.9%) were critical type.
The most common symptoms of COVID-19 in elderly patients were fever (81.0%), cough (59.9%), fatigue (42.2%), dyspnea (40.8%), and expectoration (32.0%). However, diarrhea (10.2%), pharyngalgia (4.8%), nausea (4.1%), vomiting (4.1%), and myalgia (3.4%) were rare. The median of the period from the first symptom onset to admission was 10 (IQR 7–14) days. Regarding comorbidities, 18.4% had respiratory disease (chronic bronchitis, chronic pneumonia, bronchial asthma, tuberculosis, etc.), 53.7% had cardiovascular disease (hypertension, coronary heart disease, atrial fibrillation, etc.), 21.8% had endocrine system disease (diabetes, hyperthyroidism, etc.), 3.4% had tumor, 21.8% had previously undergone surgery, and 19.1% had other comorbidities, including cirrhosis, cerebral infarction, etc. The total case-fatality rate was 28.6% (42/147, including 23 males and 19 females with a ratio of 1.21:1). The case-fatality rate increased with age, 21.8% (22/101) in patients aged 70–79 yo, 38.1% (16/42) in patients aged 80–89 yo, and 100% (4/4) in patients aged over 90 yo.
Of the entire cohort, 104 patients were cured or obviously improved until March 15, 2020. The survivors were younger (75 vs. 79, P = 0.005) and consisted of more patients in general type (P = 0.001; Table 1).
Laboratory Findings
Regarding laboratory findings at admission to the hospital (Table 2; Additional file 2), the levels of leukocyte (8.16 vs. 5.96, P = 0.003) and neutrophils (7.03 vs. 4.26, P < 0.001) increased significantly in the non-survivors. However, the level of lymphocytes (0.55 vs. 0.88, P < 0.001) decreased more significantly in the non-survivors. Elevated level of aspartate aminotransferase (AST; P = 0.007), lactic dehydrogenase (LDH; P < 0.001), creatine kinase (P = 0.034), creatinine (P = 0.033), and blood urea nitrogen (BUN; P = 0.001) were also observed in non-survivors of COVID-19 patients. In addition, indicators of inflammation, bacterial infection and blood coagulation, including procalcitonin (0.166 ng/mL vs. 0.076 ng/mL, P < 0.001), C-reactive protein (CRP; 107.3 vs. 37.65, P < 0.001), and D-dimer (6.32 vs. 1.09, P < 0.001), showed a higher level in the non-survivors.
Compared to the survivors, levels of CD3, CD4, CD8, CD19, and CD16+CD56 T cells were decreased significantly in the non-survivors, while the level of immunoglobulin (Ig)A and IgE were increased (Table 2). Differences in the other indicators of humoral immunity, including IgG, IgM, Complement component 3 and 4, were not significant between the two groups (Table 2).
Treatments in Elderly Patients with COVID-19
The main treatments included antiviral therapy (arbidol, oseltamivir, ribavirin, etc.; 97.3%), antimicrobial therapy (moxifloxacin, cefoperazone, meropenem, etc.; 84.4%), oxygen therapy (89.8%), and traditional Chinese medicine (73.5%). According to the individual’s health conditions, hormones (47.6%), gamma globulin (44.9%) and vasoactive drugs (21.8%) were also used as personalized medicine. At the same time, according to the comorbidities of COVID-19 patients, hemodialysis (2.6%) and other corresponding symptomatic support treatments, including transfusion of human albumin, nutrition support and so on, were provided (Table 3).
As is shown in Table 3, antibiotic treatment (P < 0.001) and vasoactive drugs (P < 0.001) were used more often in the non-survivors. Further studies are still required to confirm the definitive correlations between treatments and outcomes.
Predictors for Death of Elderly Patients with COVID-19
Kaplan-Meier curves indicated that severe and critical type, and elevated level of BUN increased the risk of death in elderly patients with COVID-19 (P = 0.00075 and < 0.0001 respectively by log-rank test; Figure 1). Cox proportional hazards regression were performed to identify the risk factors that were associated with the outcomes of COVID-19 patients. As summarized in Table 4, factors including age, type, level of leukocyte, neutrophils, lymphocytes, AST, CK, BUN, LDH, procalcitonin, CD3, CD4, CD8, CD16+CD56 T cells, hormone therapy, antiviral therapy, and vasoactive drugs were associated with the outcomes of elderly patients. In multivariate Cox proportional hazard regression analyses, clinical subtypes including the severe type (HR = 2.983, 95%CI: 1.231–7.226, P = 0.016), and the critical type (HR = 3.267, 95%CI: 1.009–10.576, P = 0.048) were associated with increasing risk of death when compared with the general type, and BUN greater than 9.5 mmol/L (HR = 2.805, 95% CI: 1.141–6.892, P = 0.025) on admission was the only risk factor for death among laboratory findings.