High serum lactate dehydrogenase and dyspnea: Positive predictors of adverse outcome in critical COVID-19 patients in Yichang

Comparison of inflammatory markers, coagulation indicators and outcomes between influenza and COVID-19 infection amongst children: A systematic review and meta-analysis

Background

Influenza and COVID-19 patients share similar features and outcomes amongst adults. However, the difference between these diseases is not explored in paediatric age group especially in terms of inflammatory markers, coagulation profile and outcomes. Hence, we did this review to compare the inflammatory, coagulation features and outcomes between influenza and COVID-19 infected children.

Methods

Literature search was done in PubMed Central, Scopus, EMBASE, CINAHL, Cochrane library, Google Scholar & ScienceDirect from November 2019 to May 2022. Risk of bias assessment was done through Newcastle Ottawa scale. Meta-analysis was done using random-effects model and the final pooled estimate was reported as pooled odds ratio (OR) or standardized mean difference (SMD) along with 95 % confidence interval (CI) depending on the type of outcome.

Results

About 16 studies were included with most studies having higher risk of bias. Influenza paediatric patients had significantly higher erythrocyte sedimentation rate (ESR) (pooled SMD = 0.60; 95%CI: 0.30-0.91; I2 = 0 %), lactate dehydrogenase (LDH) (pooled SMD = 2.01; 95%CI: 0.37-3.66; I2 = 98.4 %) and prothrombin time (PT) (pooled SMD = 2.12; 95%CI: 0.44-3.80; I2 = 98.3 %) when compared to paediatric COVID-19 patients. There was no significant difference in terms of features like CRP, procalcitonin, serum albumin, aPTT, mortality and need for mechanical ventilation.

Conclusion

Inflammatory markers like ESR, LDH and PT was significantly higher in influenza patients when compared to COVID-19 in children, while rest of the markers and adverse clinical outcomes were similar between both the groups. Identification of these biomarkers has helped in understanding the distinctness of COVID-19 and influenza virus and develop better management strategies.

Prediction of oxygen supplementation by a deep-learning model integrating clinical parameters and chest CT images in COVID-19

PURPOSE

As of March 2023, the number of patients with COVID-19 worldwide is declining, but the early diagnosis of patients requiring inpatient treatment and the appropriate allocation of limited healthcare resources remain unresolved issues. In this study we constructed a deep-learning (DL) model to predict the need for oxygen supplementation using clinical information and chest CT images of patients with COVID-19.

MATERIALS AND METHODS

We retrospectively enrolled 738 patients with COVID-19 for whom clinical information (patient background, clinical symptoms, and blood test findings) was available and chest CT imaging was performed. The initial data set was divided into 591 training and 147 evaluation data. We developed a DL model that predicted oxygen supplementation by integrating clinical information and CT images. The model was validated at two other facilities (n = 191 and n = 230). In addition, the importance of clinical information for prediction was assessed.

RESULTS

The proposed DL model showed an area under the curve (AUC) of 89.9% for predicting oxygen supplementation. Validation from the two other facilities showed an AUC > 80%. With respect to interpretation of the model, the contribution of dyspnea and the lactate dehydrogenase level was higher in the model.

CONCLUSIONS

The DL model integrating clinical information and chest CT images had high predictive accuracy. DL-based prediction of disease severity might be helpful in the clinical management of patients with COVID-19.

Hybrid feature engineering of medical data via variational autoencoders with triplet loss: a COVID-19 prognosis study

Medical machine learning frameworks have received much attention in recent years. The recent COVID-19 pandemic was also accompanied by a surge in proposed machine learning algorithms for tasks such as diagnosis and mortality prognosis. Machine learning frameworks can be helpful medical assistants by extracting data patterns that are otherwise hard to detect by humans. Efficient feature engineering and dimensionality reduction are major challenges in most medical machine learning frameworks. Autoencoders are novel unsupervised tools that can perform data-driven dimensionality reduction with minimum prior assumptions. This study, in a novel approach, investigated the predictive power of latent representations obtained from a hybrid autoencoder (HAE) framework combining variational autoencoder (VAE) characteristics with mean squared error (MSE) and triplet loss for forecasting COVID-19 patients with high mortality risk in a retrospective framework. Electronic laboratory and clinical data of 1474 patients were used in the study. Logistic regression with elastic net regularization (EN) and random forest (RF) models were used as final classifiers. Moreover, we also investigated the contribution of utilized features towards latent representations via mutual information analysis. HAE Latent representations model achieved decent performance with an area under ROC curve of 0.921 (±0.027) and 0.910 (±0.036) with EN and RF predictors, respectively, over the hold-out data in comparison with the raw (AUC EN: 0.913 (±0.022); RF: 0.903 (±0.020)) models. The study aims to provide an interpretable feature engineering framework for the medical environment with the potential to integrate imaging data for efficient feature engineering in rapid triage and other clinical predictive models.

Risk Factors Associated With Six-Month Mortality in Hospitalized COVID-19 Patients: A Single-Institution Study

Background Coronavirus disease 2019 (COVID-19) infection can vary from asymptomatic infection to multi-organ dysfunction. The most serious complication of infection with COVID-19 is death. Various comorbid conditions and inflammatory markers have been associated with an increased risk of mortality, specifically within the immediate post-infection period; however, less is known about long-term mortality outcomes. Objectives Our objective is to determine risk factors associated with six-month mortality in hospitalized COVID-19 patients. Methods This is a single-institution, retrospective study. We included patients hospitalized with COVID-19 from the University of Toledo Medical Center in Toledo, Ohio, who were admitted from March 20, 2020, to June 30, 2021. This study was approved by a biomedical institutional review board at the University of Toledo. Patients with available pre-stored blood samples for laboratory testing were included, and hospital charts were assessed up to six months from the date of a positive COVID-19 test result. Two groups were created based on the mortality outcome at six months from COVID-19 positive test results: survivors and non-survivors. The clinical variables or outcomes and laboratory values were compared between the two groups using non-parametric methods due to the small sample size and non-normality of the data. Either the Mann-Whitney U-test for continuous variables or Fisher's exact test for categorical variables was used for statistical analysis. Results Lactate dehydrogenase (LDH) and D-dimer levels on admission were found to be significantly higher in non-survivors than in survivors. The median high D-dimer level in non-survivors was 5.96 micrograms/milliliter (μg/mL) (interquartile range (IQR): 3.95-11.29 μg/mL) vs 1.82 μg/mL (IQR 1.13-5.55 μg/mL) in survivors (p = 0.019). Median LDH levels were also higher in non-survivors vs survivors, i.e., 621.00 international units per liter (IU/L) (IQR 440.00-849.00 IU/L) vs 328.00 IU/L (IQR 274.00-529.00 IU/L), respectively (p = 0.032). The demographic profile, comorbidity profile, and laboratory data (typically associated with short-term mortality, inflammation, and organ dysfunction) were similar between survivors and non-survivors, except for LDH and D-dimer. Conclusion Higher LDH and D-dimer levels on admission were found to be associated with an increased six-month mortality rate in hospitalized COVID-19 patients. These hematologic data can serve as risk stratification tools to prevent long-term mortality outcomes and provide proactive clinical care in hospitalized COVID-19 patients.

Utility of IL-6 in the Diagnosis, Treatment and Prognosis of COVID-19 Patients: A Longitudinal Study

COVID-19 has caused devastating effects worldwide ever since its origin in December 2019. IL-6 is one of the chief markers used in the management of COVID-19. We conducted a longitudinal study to investigate the role of IL-6 in diagnosis, treatment, and prognosis of COVID-19-related cytokine storm. Patients with COVID-19 who were admitted at AIIMS Rishikesh from March to December 2020 were included in the study. Patients with no baseline IL-6 value at admission and for whom clinical data were not available were excluded. Clinical and laboratory data of these patients were collected from the e-hospital portal and entered in an excel sheet. Correlation was seen with other inflammatory markers and outcomes were assessed using MS Excel 2010 and SPSS software. A total of 131 patients were included in the study. Of these, 74.8% were males, with mean age 55.03 ± 13.57 years, and mean duration from symptom onset being 6.69 ± 6.3 days. A total of 82.4% had WHO severe category COVID-19, with 46.56% having severe hypoxia at presentation and 61.8% of them having some comorbidity. Spearman rank correlation coefficient of IL-6 with D-dimer was 0.203, with LDH was -0.005, with ferritin was 0.3, and with uric acid was 0.123. A total of 11 patients received Tocilizumab at a mean duration from symptom onset of 18.09 days, and 100% mortality was observed. Deaths were reported more in the group with IL-6 ≥ 40 pg/mL (57.1% vs. 40.2%, p = 0.06). ICU admissions and ventilator requirement were higher in the IL-6 ≥ 40 pg/mL group (95.9% vs. 91.4%, p = 0.32 and 55.1% vs. 37.8%, p = 0.05). The study showed that IL-6 can be used as a possible "thrombotic cytokine marker". Higher values of IL-6 (≥40 pg/mL) are associated with more deaths, ICU admissions, and ventilator requirement.

Routine laboratory parameters, including complete blood count, predict COVID-19 in-hospital mortality in geriatric patients

To reduce the mortality of COVID-19 older patients, clear criteria to predict in-hospital mortality are urgently needed. Here, we aimed to evaluate the performance of selected routine laboratory biomarkers in improving the prediction of in-hospital mortality in 641 consecutive COVID-19 geriatric patients (mean age 86.6 ± 6.8) who were hospitalized at the INRCA hospital (Ancona, Italy). Thirty-four percent of the enrolled patients were deceased during the in-hospital stay. The percentage of severely frail patients, assessed with the Clinical Frailty Scale, was significantly increased in deceased patients compared to the survived ones. The age-adjusted Charlson comorbidity index (CCI) score was not significantly associated with an increased risk of death. Among the routine parameters, neutrophilia, eosinopenia, lymphopenia, neutrophil-to-lymphocyte ratio (NLR), C-reactive protein, procalcitonin, IL-6, and NT-proBNP showed the highest predictive values. The fully adjusted Cox regressions models confirmed that high neutrophil %, NLR, derived NLR (dNLR), platelet-to-lymphocyte ratio (PLR), and low lymphocyte count, eosinophil %, and lymphocyte-to-monocyte ratio (LMR) were the best predictors of in-hospital mortality, independently from age, gender, and other potential confounders. Overall, our results strongly support the use of routine parameters, including complete blood count, in geriatric patients to predict COVID-19 in-hospital mortality, independent from baseline comorbidities and frailty.

Charismatic Trends in COVID-19 Patients in Pakistan: A Case Series

Severe acute respiratory coronavirus-2 syndrome (SARS-CoV-2), the novel coronavirus causing the coronavirus disease (COVID-19), spread across the world, resulting in a global crisis. This pandemic has caused consequences that are beyond the boundaries of a single discipline of life, but it is healthcare that is under the most stress. As we received COVID-19 cases in our hospital (a private tertiary care facility in Sialkot, Pakistan), we geared up to accommodate these cases, since the government sector was already overburdened. The purpose of this study is to report the trends observed in 80 COVID-19 patients admitted at our facility from May 16 to July 14, 2020.

Comparative evaluation of semi-quantitative CT-severity scoring versus serum lactate dehydrogenase as prognostic biomarkers for disease severity and clinical outcome of COVID-19 patients

Background

Coronavirus disease 2019 pandemic causes significant strain on healthcare infrastructure and medical resources. So, it becomes crucial to identify reliable predictor biomarkers for COVID-19 disease severity and short term mortality. Many biomarkers are currently investigated for their prognostic role in COVID-19 patients. Our study is retrospective and aims to evaluate role of semi-quantitative CT-severity scoring versus LDH as prognostic biomarkers for COVID-19 disease severity and short-term clinical outcome.

Results

Two hundred sixty-six patients between April 2020 and November 2020 with positive RT-PCR results underwent non-enhanced CT scan chest in our hospital and were retrospectively evaluated for CT severity scoring and serum LDH level measurement. Data were correlated with clinical disease severity. CT severity score and LDH were significantly higher in severe and critical cases compared to mild cases (P value < 0.001). High predictive significance of CT severity score for COVID-19 disease course noted, with cut-off value ≥ 13 highly predictive of severe disease (96.96% accuracy); cut-off value ≥ 16 highly predictive of critical disease (94.21% accuracy); and cut-off value ≥ 19 highly predictive of short-term mortality (92.56% accuracy). CT severity score has higher sensitivity, specificity, positive, and negative predictive values as well as overall accuracy compared to LDH level in predicting severe, critical cases, and short-term mortality.

Conclusion

Semi-quantitative CT severity scoring has high predictive significance for COVID-19 disease severity and short-term mortality with higher sensitivity, specificity, and overall accuracy compared to LDH. Our study strongly supports the use of CT severity scoring as a powerful prognostic biomarker for COVID-19 disease severity and short-term clinical outcome to allow triage of need for hospital admission, earlier medical interference, and to effectively prioritize medical resources for cases with high mortality risk for better decision making and clinical outcome.

A machine learning based exploration of COVID-19 mortality risk

Early prediction of patient mortality risks during a pandemic can decrease mortality by assuring efficient resource allocation and treatment planning. This study aimed to develop and compare prognosis prediction machine learning models based on invasive laboratory and noninvasive clinical and demographic data from patients' day of admission. Three Support Vector Machine (SVM) models were developed and compared using invasive, non-invasive, and both groups. The results suggested that non-invasive features could provide mortality predictions that are similar to the invasive and roughly on par with the joint model. Feature inspection results from SVM-RFE and sparsity analysis displayed that, compared with the invasive model, the non-invasive model can provide better performances with a fewer number of features, pointing to the presence of high predictive information contents in several non-invasive features, including SPO2, age, and cardiovascular disorders. Furthermore, while the invasive model was able to provide better mortality predictions for the imminent future, non-invasive features displayed better performance for more distant expiration intervals. Early mortality prediction using non-invasive models can give us insights as to where and with whom to intervene. Combined with novel technologies, such as wireless wearable devices, these models can create powerful frameworks for various medical assignments and patient triage.

Serum lactate dehydrogenase activities as systems biomarkers for 48 types of human diseases

Most human diseases are systems diseases, and systems biomarkers are better fitted for diagnostic, prognostic, and treatment monitoring purposes. To search for systems biomarker candidates, lactate dehydrogenase (LDH), a housekeeping protein expressed in all living cells, was investigated. To this end, we analyzed the serum LDH activities from 172,933 patients with 48 clinically defined diseases and 9528 healthy individuals. Based on the median values, we found that 46 out of 48 diseases, leading by acute myocardial infarction, had significantly increased (p < 0.001), whereas gout and cerebral ischemia had significantly decreased (p < 0.001) serum LDH activities compared to the healthy control. Remarkably, hepatic encephalopathy and lung fibrosis had the highest AUCs (0.89, 0.80), sensitivities (0.73, 0.56), and specificities (0.90, 0.91) among 48 human diseases. Statistical analysis revealed that over-downregulation of serum LDH activities was associated with blood-related cancers and diseases. LDH activities were potential systems biomarker candidates (AUCs > 0.8) for hepatic encephalopathy and lung fibrosis.

Predictors of in-hospital mortality AND death RISK STRATIFICATION among COVID-19 PATIENTS aged ≥ 80 YEARs OLD

Introduction

: To date, mainly due to age-related vulnerability and to coexisting comorbidities, older patients often face a more severe COVID-19. This study aimed to identify at Emergency Department (ED) admission the predictors of in-hospital mortality and suitable scores for death risk stratification among COVID-19 patients ≥ 80 years old.

Methods

: Single-centre prospective study conducted in the ED of an university hospital, referral center for COVID-19 in central Italy. We included 239 consecutive patients ≥ 80 years old with laboratory-confirmed COVID-19.

The primary study endpoint was all-cause in-hospital mortality. Multivariable Cox regression analysis was performed on significant variables at univariate analysis to identify independent risk factor for death. Overall performance in predicting mortality of WHO severity scale, APACHE II score, NEWS score, and CURB-65 was calculated.

Results

: Median age was 85 [82-89] and 112 were males (46.9%). Globally, 77 patients (32.2%) deceased. The presence of consolidations at chest x-ray and the hypoxemic respiratory failure were significant predictors of poor prognosis. Moreover, age ≥ 85 years, dependency in activities of daily living (ADL), and dementia were risk factors for death, even after adjusting for clinical covariates and disease severity. All the evaluated scores showed a fairly good predictive value in identifying patients who could experience a worse outcome.

Conclusions

: Among patients ≥ 80 years old hospitalized with COVID-19, not only a worse clinical and radiological presentation of the disease, but also the increasing age, dementia, and impairment in ADL were strong risk factors for in-hospital death, regardless of disease severity.