Homogeneous ensemble models for predicting infection levels and mortality of COVID-19 patients: Evidence from China

Background Persistence of long-term COVID-19 pandemic is putting high pressure on healthcare services worldwide for several years. This article aims to establish models to predict infection levels and mortality of COVID-19 patients in China. Methods Machine learning models and deep learning models have been built based on the clinical features of COVID-19 patients. The best models are selected by area under the receiver operating characteristic curve (AUC) scores to construct two homogeneous ensemble models for predicting infection levels and mortality, respectively. The first-hand clinical data of 760 patients are collected from Zhongnan Hospital of Wuhan University between 3 January and 8 March 2020. We preprocess data with cleaning, imputation, and normalization. Results Our models obtain AUC = 0.7059 and Recall (Weighted avg) = 0.7248 in predicting infection level, while AUC=0.8436 and Recall (Weighted avg) = 0.8486 in predicting mortality ratio. This study also identifies two sets of essential clinical features. One is C-reactive protein (CRP) or high sensitivity C-reactive protein (hs-CRP) and the other is chest tightness, age, and pleural effusion. Conclusions Two homogeneous ensemble models are proposed to predict infection levels and mortality of COVID-19 patients in China. New findings of clinical features for benefiting the machine learning models are reported. The evaluation of an actual dataset collected from January 3 to March 8, 2020 demonstrates the effectiveness of the models by comparing them with state-of-the-art models in prediction.

Predicting Progression of COVID-19 Infection to Prioritize Medical Resource Allocation: A Novel Triage Model Based on Patient Characteristics and Symptoms at Presentation.

Background: The COVID-19 global pandemic has posed unprecedented challenges to health care systems all over the world. The speed of the viral spread results in a tsunami of patients, which begs for a reliable screening tool using readily available data to predict disease progression. Methods: Multicenter retrospective cohort study was performed to develop and validate a triage model. Patient demographic and non-laboratory clinical data were recorded. Using only the data from Zhongnan Hospital, step-wise multivariable logistic regression was performed, and a prognostic nomogram was constructed based on the independent variables identifies. The discrimination and calibration of the model were validated. External independent validation was performed to further address the utility of this model using data from Jinyintan Hospital. Results: A total of 716 confirmed COVID-19 cases from Zhongnan Hospital were included for model construction. Men, increased age, fever, hypertension, cardio-cerebrovascular disease, dyspnea, cough, and myalgia are independent risk factors for disease progression. External independent validation was carried out in a cohort with 201 cases from Jinyintan Hospital. The area under the curve (AUC) was 0.787 (95% confidence interval [CI]: 0.747-0.827) in the training group and 0.704 (95% CI: 0.632-0.777) in the validation group. Conclusions: We developed a novel triage model based on basic and clinical data. Our model could be used as a pragmatic screening aid to allow for cost efficient screening to be carried out such as over the phone, which may reduce disease propagation through limiting unnecessary contact. This may help allocation of limited medical resources.

Prevalence and Characteristics of Hypoxic Hepatitis in COVID-19 Patients in the Intensive Care Unit: A First Retrospective Study.

Purpose: Coronavirus disease 2019 (COVID-19) has been associated with acute liver injury in reports worldwide. But no studies to date have described hypoxic hepatitis (HH) in patients with COVID-19. We aim to identify the prevalence of and possible mechanisms of HH in COVID-19 patients in the Intensive Care Unit (ICU). Methods: This retrospective study was conducted on 51 patients with confirmed SARS-CoV-2 infection in the ICU at Zhongnan Hospital of Wuhan University from December 21, 2019, to March 11, 2020. Information on clinical features of enrolled patients was collected for analysis. Results: HH was observed in 5.88% of the ICU patients with SARS-CoV-2 infection. All HH patients were progressing to respiratory failure and peak alanine aminotransferase (ALT) values were 1665, 1414, and 1140 U/L during hospitalization, respectively. All patients with HH died as a result of the deterioration of multiple organ failure (MOF). The dynamic changes of ALT, aspartate transaminase (AST), and total bilirubin (TBIL) levels were more dramatic in HH groups. Levels of TBIL, C-reactive protein (CRP), procalcitonin (PCT), and interleukin-6(IL-6) showed statistically significant elevation in HH cases compared with that in non-HH cases (P < 0.001). Besides, the median survival time of the HH group was significantly shorter than the non-HH group (P < 0.05). Conclusions: In ICU, HH was not a rare condition in patients with severe COVID-19 and has a high mortality. The main causes of HH are respiratory and cardiac failure and may be associated with the immune-mediated inflammatory response. Clinicians should search for any underlying hemodynamic or respiratory instability even in patients with normal ALT levels on admission.

Cancer patients in SARS-CoV-2 infection: a single-center experience from Wuhan.

Background: The Coronavirus disease 2019 (COVID-19) global pandemic has posed unprecedented challenges to the health-care systems all over the world. Among the booming literatures about COVID-19, there is yet a paucity of study addressing the association between COVID-19 and cancer, which is a rare comorbidity of COVID-19, as well as consensus for treatment of cancer in this pandemic. Methods: In this retrospective, single-center cohort study, information of all inpatient cases with laboratory-confirmed COVID-19 who had treatment outcome were collected from the designated departments in Zhongnan Hospital of Wuhan University, Wuhan, China on March 10, 2020. Demographic data, clinical information, and treatment outcomes were extracted from electronic medical records. Severe events were defined as admission to intensive care unit (ICU), the use of mechanical ventilation, or death. Result: A total of 716 patients with laboratory-confirmed COVID-19 infection were identified. Among them, a total of 12 cases (1.7%, 95% CI: 0.7%-2.6%) had history of cancer with 4 cases (33%) experienced severe events. Compared with cases without cancer, patients with cancer have higher risks of severe events (33% vs 7.7%, p=0.012) and deaths (25% vs 3.6%, p=0.009). Multivariable logistic regression model showed that cancer was independently associated with increased odds of severe events after adjusting for other risk factors (OR 6.51, 95% CI 1.72-24.64; p=0.006). Among COVID-19 patients with cancer, we found that patients older than 60 years (75%), with other comorbidities (50%), or experiencing anticancer treatment in past month (42.9%) had a numerically higher incidence of severe events. Conclusion: Cancer is a rare comorbidity of patients with COVID-19; however, it cannot be overemphasized due to its poorer outcomes. We propose that personalized treatment recommendation for cancer patients should be addressed during COVID-19 pandemic, along with meticulous personal protective protocols for them to mitigate the risk of SARS-CoV-2 infection.

The association between markers of liver injury and clinical outcomes in patients with COVID-19 in Wuhan.

BACKGROUND: The outbreak of coronavirus disease 2019 (COVID-19) is a critical challenge for public health. The effect of COVID-19 on liver injury has not been fully established. AIMS: To evaluate the dynamic changes in liver function and the relationship between liver damage and prognosis in patients with COVID-19. METHODS: Retrospective analysis of clinical data of 675 patients with COVID-19 in Zhongnan Hospital of Wuhan University from January 3 to March 8, 2020. Patients were classified as having normal or abnormal liver function and liver injury. RESULTS: Of 675 patients, 253 (37.5%) had abnormal liver function during hospitalisation, and 52 (7.7%) had liver injury. The dynamic changes of ALT and AST levels were more significant in patients with liver injury and in those who died. AST >3-fold upper limit of normal (ULN) had the highest risk of death and mechanical ventilation. Compared to patients with normal AST levels, mortality and risk of mechanical ventilation significantly increased 19.27-fold (95% confidence interval [CI], 4.89-75.97; P < 0.0001) and 116.72-fold (95% CI, 31.58-431.46; P < 0.0001), respectively, in patients with AST above 3-fold ULN. Increased leucocytes, decreased lymphocytes and female sex were independently associated with liver injury. CONCLUSIONS: The dynamic changes in liver function may have a significant correlation with the severity and prognosis of COVID-19. Increased index of liver injury was closely related to mortality and need for mechanical ventilation. Therefore, these indicators should be closely monitored during hospitalisation.