Deep-learning artificial intelligence analysis of clinical variables predicts mortality in COVID-19 patients.

Objective: The large number of clinical variables associated with coronavirus disease 2019 (COVID-19) infection makes it challenging for frontline physicians to effectively triage COVID-19 patients during the pandemic. This study aimed to develop an efficient deep-learning artificial intelligence algorithm to identify top clinical variable predictors and derive a risk stratification score system to help clinicians triage COVID-19 patients. Methods: This retrospective study consisted of 181 hospitalized patients with confirmed COVID-19 infection from January 29, 2020 to March 21, 2020 from a major hospital in Wuhan, China. The primary outcome was mortality. Demographics, comorbidities, vital signs, symptoms, and laboratory tests were collected at initial presentation, totaling 78 clinical variables. A deep-learning algorithm and a risk stratification score system were developed to predict mortality. Data were split into 85% training and 15% testing. Prediction performance was compared with those using COVID-19 severity score, CURB-65 score, and pneumonia severity index (PSI). Results: Of the 181 COVID-19 patients, 39 expired and 142 survived. Five top predictors of mortality were D-dimer, O2 Index, neutrophil:lymphocyte ratio, C-reactive protein, and lactate dehydrogenase. The top 5 predictors and the resultant risk score yielded, respectively, an area under curve (AUC) of 0.968 (95% CI = 0.87-1.0) and 0.954 (95% CI = 0.80-0.99) for the testing dataset. Our models outperformed COVID-19 severity score (AUC = 0.756), CURB-65 score (AUC = 0.671), and PSI (AUC = 0.838). The mortality rates for our risk stratification scores (0-5) were 0%, 0%, 6.7%, 18.2%, 67.7%, and 83.3%, respectively. Conclusions: Deep-learning prediction model and the resultant risk stratification score may prove useful in clinical decisionmaking under time-sensitive and resource-constrained environment.

Comparison of deep vein thrombosis risks in acute respiratory distress syndrome caused by COVID-19 and bacterial pneumonia: a retrospective cohort study.

BACKGROUND: High incidence of deep vein thrombosis (DVT) has been observed in patients with acute respiratory distress syndrome (ARDS) caused by COVID-19 and those by bacterial pneumonia. However, the differences of incidence and risk factors of DVT in these two groups of ARDS had not been reported before. STUDY DESIGN AND METHODS: We performed a retrospective cohort study to investigate the difference of DVT in incidence and risk factors between the two independent cohorts of ARDS and eventually enrolled 240 patients, 105 of whom with ARDS caused by COVID-19 and 135 caused by bacterial pneumonia. Lower extremity venous compression ultrasound scanning was performed whenever possible regardless of clinical symptoms in the lower limbs. Clinical characteristics, including demographic information, clinical history, vital signs, laboratory findings, treatments, complications, and outcomes, were analyzed for patients with and without DVT in these two cohorts. RESULTS: The 28-days incidence of DVT was higher in patients with COVID-19 than in those with bacterial pneumonia (57.1% vs 41.5%, P = 0.016). Taking death as a competitive risk, the Fine-Gray test showed no significant difference in the 28-day cumulative incidence of DVT between these two groups (P = 0.220). Fine-Gray competing risk analysis also showed an association between increased CK (creatine kinase isoenzyme)-MB levels (P = 0.003), decreased PaO2 (partial pressure of arterial oxygen)/FiO2 (fraction of inspired oxygen) ratios (P = 0.081), increased D-dimer levels (P = 0.064) and increased incidence of DVT in COVID-19 cohort, and an association between invasive mechanical ventilation (IMV; P = 0.001) and higher incidence of DVT and an association between VTE prophylaxis (P = 0.007) and lower incidence of DVT in bacterial pneumonia cohort. The sensitivity and specificity of the corresponding receiver operating characteristic curve originating from the combination of CK-MB levels, PaO2/FiO2 ratios, and D-dimer levels ≥0.5 µg/mL were higher than that of the DVT Wells score (P = 0.020) and were not inferior to that of the Padua prediction score (P = 0.363) for assessing the risk of DVT in COVID-19 cohort. CONCLUSIONS: The incidence of DVT in patients with ARDS caused by COVID-19 is higher than those caused by bacterial pneumonia. Furthermore, the risk factors for DVT are completely different between these two ARDS cohorts. It is suggested that COVID-19 is probably an additional risk factor for DVT in ARDS patients.

The Gaps Between the Self and Professional Evaluation in Mental Health Assessment of COVID-19 Cluster Cases.

Objective: To analyze the discrepancy between self-rating and professional evaluation of mental health status in coronavirus disease 2019 (COVID-19) cluster cases. Method: A total of 65 COVID-19 cluster cases admitted to Beijing Ditan Hospital Capital Medical University from June 14, 2020 to June 16, 2020 were included in the study. Mental health assessment was completed by self-rating and professional evaluation. The gaps between self-rating and professional evaluation in different demographic characteristics were compared. Results: The results of self-rating were inconsistent with those of professional evaluation. The gap was statistically different among certain demographic subgroups. As for anxiety, the gaps had remarkable statistics differences in subgroups of sex, monthly income, infection way, and anxiety/depression medical history. Similarly, in the terms of depression, the gaps had significant statistic differences in the subgroups of the medical history of anxiety/depression, history of physical disease, employment status and the insurance type, marriage, education (year), residing in Beijing (year), and the monthly income. Conclusion: Compared to the professional evaluation, patients had a higher self-rating, which may be related to some demographic characteristics. It suggests that screening can be conducted in patients with COVID-19 by self-rating first, and then professional evaluation should be carried out in the patients with suspicious or positive results.

Correlation between white blood cell count at admission and mortality in COVID-19 patients: a retrospective study.

BACKGROUND: Coronavirus disease-19 (COVID-19) has become a world health threaten. Its risk factors with death were still not known. White blood cells (WBC) count as a reflection of inflammation has played a vital role in COVID-19, however its level with death is not yet investigated. METHODS: In this retrospective, single-center study, all confirmed patients with COVID-19 at West Branch of Union Hospital from Jan 29 to Feb 28, 2020 were collected and analyzed. Demographic and clinical data including laboratory examinations were analyzed and compared between recovery and death patients. RESULTS: A total of 163 patients including 33 death cases were included in this study. Significant association was found between WBC count and death (HR = 1.14, 95%CI: 1.09-1.20, p < 0.001). The regression analysis results showed there was a significant association between WBC count and death (HR = 5.72, 95%CI: 2.21-14.82, p < 0.001) when use the second quartile as a cutoff value (> 6.16 × 10^9/L). The difference was still exist after adjusting for confounding factors (HR = 6.26, 95%CI: 1.72-22.77, p = 0.005). In addition, Kaplan-meier survival analysis showed that there was a significant decline of the cumulative survival rate (p < 0.001) in those with WBC count ≥6.16 × 10^9/L. CONCLUSION: WBC count at admission is significantly corelated with death in COVID-19 patients. Higher level of WBC count should be given more attention in the treatment of COVID-19.

Scores based on neutrophil percentage and lactate dehydrogenase with or without oxygen saturation predict hospital mortality risk in severe COVID-19 patients.

BACKGROUND: Risk scores are needed to predict the risk of death in severe coronavirus disease 2019 (COVID-19) patients in the context of rapid disease progression. METHODS: Using data from China (training dataset, n = 96), prediction models were developed by logistic regression and then risk scores were established. Leave-one-out cross validation was used for internal validation and data from Iran (test dataset, n = 43) was used for external validation. RESULTS: A NSL model (area under the curve (AUC) 0.932) and a NL model (AUC 0.903) were developed based on neutrophil percentage and lactate dehydrogenase with and without oxygen saturation (SaO2) using the training dataset. AUCs of the NSL and NL models in the test dataset were 0.910 and 0.871, respectively. The risk scoring systems corresponding to these two models were established. The AUCs of the NSL and NL scores in the training dataset were 0.928 and 0.901, respectively. At the optimal cut-off value of NSL score, the sensitivity and specificity were 94% and 82%, respectively. The sensitivity and specificity of NL score were 94% and 75%, respectively. CONCLUSIONS: These scores may be used to predict the risk of death in severe COVID-19 patients and the NL score could be used in regions where patients' SaO2 cannot be tested.

High prevalence of occult thrombosis in cases of mild/moderate COVID-19.

BACKGROUND AND PURPOSE: An increasing number of reports have observed thrombosis in severe cases of COVID-19. The aim of this study was to evaluate the incidence of thromboembolism in mild/moderate cases of COVID-19. All of the patients had normal coagulation tests and none had any overt thrombotic complications. Our findings indicate that it is important to screen the thrombotic status of cases with mild/moderate COVID-19. METHODS: Between 11 June and 8 July 2020, 23 patients with mild/moderate COVID-19 pneumonia consented to having computed tomography pulmonary angiography (CPTA) and computed tomography venography (CTV) scans of the lungs and extremity veins. Doppler ultrasound (DUS) was also performed in all patients for screening. The incidence, clinical manifestations, laboratory examinations, imaging features, and prognosis, of patients with venous thromboembolism (VTE) were analyzed and compared with those of patients with COVID-19 pneumonia without VTE. RESULTS: Nineteen patients (82.6%) had VTE, mainly distal limb thrombosis. Only one of the VTE patients was positive when screened by DUS; the other VTE patients were negative by DUS. All of the mild/moderate patients with VTE were screened by CTPA + CTV. Blood tests for inflammatory, coagulation, and biochemical, parameters were all within the normal range, except for WBC and LDH. CONCLUSIONS: When using CTV screening for DVT, we found that the incidence of thrombosis in patients with mild/moderate COVID-19 markedly increased to 82.6% (19/23). Screening for thrombosis is therefore important in patients with COVID-19. CTV is more sensitive than DUS for the detection of thrombosis. More research is now needed to evaluate the significance of thrombosis in COVID-19 pneumonia.

The immunodominant and neutralization linear epitopes for SARS-CoV-2.

Although vaccines against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) are under development, the antigen epitopes on the virus and their immunogenicity are poorly understood. Here, we simulate the 3D structures and predict the B cell epitopes on the spike (S), envelope (E), membrane (M), and nucleocapsid (N) proteins of SARS-CoV-2 using structure-based approaches and validate epitope immunogenicity by immunizing mice. Almost all 33 predicted epitopes effectively induce antibody production, six of these are immunodominant epitopes in individuals, and 23 are conserved within SARS-CoV-2, SARS-CoV, and bat coronavirus RaTG13. We find that the immunodominant epitopes of individuals with domestic (China) SARS-CoV-2 are different from those of individuals with imported (Europe) SARS-CoV-2, which may be caused by mutations on the S (G614D) and N proteins. Importantly, we find several epitopes on the S protein that elicit neutralizing antibodies against D614 and G614 SARS-CoV-2, which can contribute to vaccine design against coronaviruses.

Risk factors associated with disease aggravation among 126 hospitalized patients with COVID-19 in different places in China: A retrospective observational study.

Coronavirus disease 2019 (COVID-19) has rapidly spread on a global scale. Therefore, it is urgent to identify risk factors that could be associated with severe type of COVID-19 from common type.For this retrospective study, we recruited patients with COVID-19 in Wuhan and Zhoukou. Patients were classified into a severe group and common group based on guidelines after admission. Clinical manifestations and laboratory tests were compared, and univariate binary logistic regression and multivariate regression analyses were applied to assess potential risk factors.A total of 126 patients were recruited from January 23 to March 23, 2020. Ninety cases were identified as the common type and 36 as the severe type. The average age in the severe group was significantly older than that in the common group (P = .008). Patients with severe COVID-19 exhibited higher proportions of dyspnea (P = .001), weakness (P = .023), and diarrhea (P = .046). Moreover, there were more patients with hypertension (P = .01) or coinfection (P = .001) in the severe group than in the common group. Additionally, severe COVID-19 was associated with increased neutrophil counts (P < .001), C-reactive protein (P < .001), procalcitonin (P = .024) and decreased lymphocyte counts (P = .001), hemoglobin (P < .001), total protein (TP) (P < .001), and albumin (ALB) (P < .001). Based on logistic regression analysis, dyspnea (P < .001), TP (P = .042), and ALB (P = .003) were independent risk factors for severe disease.Patients with lower TP, ALB, and dyspnea should be carefully monitored, and early intervention should be implemented to prevent the development of severe disease.

Deep Vein Thrombosis in Hospitalized Patients With COVID-19 in Wuhan, China: Prevalence, Risk Factors, and Outcome

BACKGROUND: To investigate deep vein thrombosis (DVT) in hospitalized patients with coronavirus disease 2019 (COVID-19), we performed a single institutional study to evaluate its prevalence, risk factors, prognosis, and potential thromboprophylaxis strategies in a large referral and treatment center. METHODS: We studied a total of 143 patients with COVID-19 from January 29, 2020 to February 29, 2020. Demographic and clinical data, laboratory data, including ultrasound scans of the lower extremities, and outcome variables were obtained, and comparisons were made between groups with and without DVT. RESULTS: Of the 143 patients hospitalized with COVID-19 (age 63±14 years, 74 [51.7%] men), 66 patients developed lower extremity DVT (46.1%: 23 [34.8%] with proximal DVT and 43 [65.2%] with distal DVT). Compared with patients who did not have DVT, patients with DVT were older and had a lower oxygenation index, a higher rate of cardiac injury, and worse prognosis, including an increased proportion of deaths (23 [34.8%] versus 9 [11.7%];P=0.001) and a decreased proportion of patients discharged (32 [48.5%] versus 60 [77.9%];P<0.001). Multivariant analysis showed an association only between CURB-65 (confusion status, urea, respiratory rate, and blood pressure) score 3 to 5 (odds ratio, 6.122;P=0.031), Padua prediction score ≥4 (odds ratio, 4.016;P=0.04), D-dimer >1.0 µg/mL (odds ratio, 5.818;P<0.014), and DVT in this cohort, respectively. The combination of a CURB-65 score 3 to 5, a Padua prediction score ≥4, and D-dimer >1.0 µg/mL has a sensitivity of 88.52% and a specificity of 61.43% for screening for DVT. In the subgroup of patients with a Padua prediction score ≥4 and whose ultrasound scans were performed >72 hours after admission, DVT was present in 18 (34.0%) patients in the subgroup receiving venous thromboembolism prophylaxis versus 35 (66.0%) patients in the nonprophylaxis group (P=0.010). CONCLUSIONS: The prevalence of DVT is high and is associated with adverse outcomes in hospitalized patients with COVID-19. Prophylaxis for venous thromboembolism may be protective in patients with a Padua protection score ≥4 after admission. Our data seem to suggest that COVID-19 is probably an additional risk factor for DVT in hospitalized patients.