Clinical Characteristics of COVID-19: Risk Factors for Early Oxygen Requirement after Hospitalization

Since December 2019, the novel severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) that causes coronavirus disease 2019 (COVID-19)1 has led to a worldwide pandemic.2 The pandemic has resulted in substantial morbidity and mortality, threatening the collapse of medical systems. The clinical course of COVID-19 varies from asymptomatic to severe or critical. Most patients have mild to moderate disease (81%), but 14% have severe disease and 5% have critical disease.3, 4, 5 Patients with mild to moderate COVID-19 can be managed well in community-treatment centers (CTCs),6, 7 but patients with severe COVID-19 require treatment at hospital settings to receive various managements, including oxygen supplementation.

Therefore, it is important to recognize severe COVID-19 in a timely manner when medical resources are insufficient. One of the most important things in the management of severe COVID-19 is supplemental oxygenation therapy.8 Some patients need oxygen therapy in the early stages of hospitalization, while others need oxygen therapy in the late stage of hospitalization. Many previous studies have identified several risk factors for severe COVID-19, such as old age, obesity, diabetes and hypertension.9, 10, 11, 12 However, the differences underlying an early oxygen requirement and late oxygen requirement in COVID-19 are still unknown, and there are still difficulties in managing patients who progress early during hospitalization. The purpose of this study was to investigate whether there are differences in the epidemiologic and clinical characteristics of COVID-19 patients with an early oxygen requirement and those with a late oxygen requirement.

During the study period, 174 COVID-19 patients who required oxygen therapy after admission were screened. A total of 164 patients were finally included in this study, excluding 10 patients who were referred by CTCs or other hospitals. Among them, 94 patients needed supplemental oxygen therapy within 3 days of hospitalization (the early group), and the other 70 patients needed supplemental oxygen therapy after 3 days (the late group) (Table 1). The two groups showed similar demographic baselines except for older age in the early group (median 73 vs. 67 years, P = 0.002).

Table 1
Baseline characteristics of coronavirus disease 2019 patients and laboratory findings at admission

• Click for larger image
• Click for full table
• Download as Excel file

Epidemiologically, the early group had a more uncertain exposure history, and the late group had a more definite exposure history, such as direct contact with confirmed patients or association with large-scale outbreaks (50% vs. 68.6%, P = 0.017). The duration from the onset of symptoms to hospitalization was significantly longer in the early group (median 5 days vs. 2 days, P < 0.001). The median duration from diagnosis to hospitalization was 0 days in both groups, indicating that most patients were hospitalized immediately after diagnosis. The median duration from onset of symptoms to initial oxygen requirement was 6 and 8 days (P < 0.001) in the early and late groups, respectively. With regard to the clinical variables at admission, the early group showed more febrile sensation (64.9% vs. 42.9%, P = 0.005), general weakness (14.9% vs. 4.3%, P = 0.037) and dyspnea (38.3% vs. 1.4%, P < 0.001), and the late group was more asymptomatic (2.1% vs. 20%, P < 0.001). The laboratory findings showed that leukocytosis and an elevation of aspartate transaminase, lactate dehydrogenase and C-reactive protein (CRP) were more prominent in the early group. Lung infiltration was present in 96.8% of the early group and in 64.3% of the late group (P < 0.001) (Table 1).

Multivariate analysis showed that older age, especially > 65 years (odds ratio [OR], 4.65), and symptoms for more than 5 days at admission (OR, 9.13) were significant predictors in the early group, and febrile sensation (OR, 6.01), dyspnea (OR, 30.0), increased CRP > 1 mg/dL (OR, 7.87) and chest X-ray abnormality (OR, 8.15) were more prominent in the early group (Table 2). Critical outcomes, such as mechanical ventilation care, extracorporeal membrane oxygenation and death, were higher in the early group (29.8% vs. 14.3%, P = 0.002), especially in the group over 65 years of age. Regarding the treatment options, the use of dexamethasone was higher in the early group (71.3% vs. 55.7%, P = 0.039), and remdesivir tended to be used more in the early group (59.6% vs. 44.3%, P = 0.052). In-hospital mortality was also higher in the early group, but the difference was not statistically significant (Table 3).

Table 2
Risk factors for the early group determined by multivariate analysis

• Click for larger image
• Click for full table
• Download as Excel file

Table 3
Clinical outcomes between the early group and the late group

• Click for larger image
• Click for full table
• Download as Excel file

As of February 2021, South Korea has maintained a strategy of prompt tracking, mass diagnosis and early isolation for most COVID-19 patients. Under these circumstances, most patients are admitted to CTCs6 or hospitals without delay, and clinicians have found that some patients become aggravated just after admission, requiring supplemental oxygen therapy, and others go through the usual clinical course of COVID-19.3, 14 In this study, the two groups had similar baseline characteristics except older age and delayed diagnosis combined with delayed admission in the early group. Other factors, such as febrile sensation, dyspnea, increased CRP and chest X-ray abnormality, were associated with this group, indicating a characteristic of severe inflammation in this group. Moreover, the early group had worse critical clinical outcomes. Timely diagnosis seems to be crucial in older people.

The strategy of rapid tracking and mass diagnostic approaches not only prevents subsequent infections but also enables COVID-19 patients to be admitted quickly for standard care.15 The patients with a definite exposure history tended to be diagnosed earlier, and their oxygen requirements were delayed. On the other hand, the early group tended to have an uncertain exposure history and was therefore not included in the active scope of tracking and diagnosis, which resulted in delayed diagnosis. There is doubt that early supportive care is beneficial in the absence of sufficiently effective specific therapeutics against COVID-19. However, our results showed that the duration from onset of illness to oxygen requirement was shorter in the early group, which suggests that early best supportive medical care is beneficial. During the study period, antibody therapy was not available, and due to the shortage, the government allowed remdesivir for patients only after the occurrence of oxygen demand. We do not use steroids in the early phase. Thus, specific therapeutics might not play a role in the improvement before the initiation of oxygen therapy. With the oxygen therapy the antiviral or anti-inflammatory treatments were more common in the early group, which might be related to the severer clinical status of the group. Clinical outcomes were also poorer, and in-hospital mortality tended to be increased in the early group.

The strategy of mass tracking and hospitalization of every case requires considerable resources, and sustainability is a continuous issue to overcome. Not every country can afford this approach. Thus, a focused approach to target older people and to provide selective tracking, diagnostic testing and timely medical care may be a reasonable option. In view of the relatively mild nature of COVID-19 in young people, this strategy may not be harmful. Our results provide positive evidence supporting a targeted approach for certain at-risk groups. Other factors may play a role in the individual differences in immune responses or immunity to COVID-19.

Our study has some limitations. First, it was a retrospective single-center study, and the number of subjects was relatively small. There were some patients with dementia from outbreaks in nursing homes who could not present their symptoms. Second, there might be some bias in selection of subjects because patients referred from CTCs or other hospitals were excluded. However, this study center is a public hospital where most patients are admitted without specific limitations. Third, other factors, such as individual differences in immune responses or age-related immunity to COVID-19, might have played a role. Basic research combined with clinical analysis may better explain the different severities of COVID-19.

In conclusion, our study showed that older age, especially >65 years, and a delay of over 5 days from the onset of illness to admission, are associated with early oxygen demand in COVID-19 patients. Interventions for earlier diagnosis of elderly people may lead to a better clinical outcome. A focused strategy of mass tracking and isolation may deserve attention in the later stage of the COVID-19 pandemic.