Search

SummaryBackgroundStarting from early December 2019, cases of human infection with a novel coronavirus were identified in Wuhan, Hubei Province, China. It spreads rapidly to other cities and numerous countries. This study was performed to investigate clinical features of patients with severe SARS-CoV-2 pneumonia and identify risk factors for converting to severe cases in those who had mild to moderate diseases.MethodsIn this retrospective, multicenter cohort study, patients with mild to moderate SARS-CoV-2 pneumonia were included from Ningbo First Hospital and Jingzhou Central Hospital. Demographic data, symptoms, laboratory values, comorbidities, and clinical outcomes were collected. Data were compared between non-severe and severe patients. Logistic regression analysis was performed to assess risk factors in predicting the patients with SARS-CoV-2 pneumonia who would convert to severe cases.Findings120 patients (36 from Ningbo First Hospital and 84 from Jingzhou Central Hospital) were included in this study, among which 62 were excluded and 58 were included in the final analysis. Compared with non-severe cases, severe patients with SARS-CoV-2 pneumonia had a longer: time to clinical recovery (12.9±4.4 vs 8.3±4.7; p=0.0011), duration of viral shedding (15.7±6.7 vs 11.8±5.0; p=0.0183), and hospital stay (20.7±1.2 vs 14.4±4.3; p=0.0211). Multivariate logistic regression indicated that lymphocyte count was significantly associated with the rate of converting to severe cases (odds ratio 1.28, 95%CI 1.06–1.54, per 0.1×109/L reduced; p=0.007), while using of low-to-moderate doses of systematic corticosteroids was associated with reduced likelihood of converting to a severe case (odds ratio 0.14, 95%CI 0.02–0.80; p=0.0275).InterpretationThe low peripheral blood lymphocyte count was an independent risk factor for SARS-CoV-2 pneumonia patients converting to severe cases. This finding may help clinicians more accurately predict prognosis, and triage priorities to improve clinical outcomes.Research in contextEvidence before this studySevere Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV2) is a novel coronavirus that have emerged in early December 2019, and has caused a novel coronavirus disease (COVID-19). It has been deemed as a public health emergency of global concern by The World Health Organization (WHO). We searched PubMed for articles published up to March 11, 2020, using the search terms (“novel coronavirus” OR “SARS-CoV-2” OR “COVID-19”) with no language or time restrictions. Previous work has described clinical characteristics of critically ill and non-critically ill patients with COVID-19. However, no published works have focused on clinical features of patients with severe SARS-CoV-2 pneumonia and identify risk factors for converting to severe cases in those who had mild to moderate diseases.Added value of this studyIn this retrospective and multicenter cohort study, we reported demographics characteristics, baseline symptoms, laboratory findings, corticosteroid usage and hospital course of patients with non-severe COVID-19 and severe COVID-19. Comparing with non-severe patients, severe patients with COVID-19 was found to have a longer: time to clinical recovery (12.9±4.4 vs 8.3±4.7; p=0.0011), duration of viral shedding (15.7±6.7 vs 11.8±5.0; p=0.0183), and hospital stay (20.7±1.2 vs 14.4±4.3; p=0.0211). By multivariate logistic regression, we found increasing odds of converting to severe cases associated with lower lymphocyte count (odds ratio 1.28, 95%CI 1.06–1.54, per 0.1×109/L reduced; p=0.007). Using of low-to-moderate doses of systematic corticosteroids was associated with reduced likelihood of converting to a severe case (odds ratio 0.14, 95%CI 0.02–0.80; p=0.0275).Implications of all the available evidenceLow lymphocyte count in peripheral blood was an independent risk factor for patients who converted to severe cases. In addition, using of systematic corticosteroids in mild to moderate patients with SARS-CoV-2 pneumonia was associated with a reduced risk of converting to severe cases. These findings may help clinicians predict prognosis more accurately, and triage priorities to improve clinical outcomes. Further prospective studies are warranted to confirm these findings.