Admission vital signs as predictors of COVID-19 mortality: a retrospective cross-sectional study

The predictive power of data: machine learning analysis for Covid-19 mortality based on personal, clinical, preclinical, and laboratory variables in a case-control study

BACKGROUND AND PURPOSE

The COVID-19 pandemic has presented unprecedented public health challenges worldwide. Understanding the factors contributing to COVID-19 mortality is critical for effective management and intervention strategies. This study aims to unlock the predictive power of data collected from personal, clinical, preclinical, and laboratory variables through machine learning (ML) analyses.

METHODS

A retrospective study was conducted in 2022 in a large hospital in Abadan, Iran. Data were collected and categorized into demographic, clinical, comorbid, treatment, initial vital signs, symptoms, and laboratory test groups. The collected data were subjected to ML analysis to identify predictive factors associated with COVID-19 mortality. Five algorithms were used to analyze the data set and derive the latent predictive power of the variables by the shapely additive explanation values.

RESULTS

Results highlight key factors associated with COVID-19 mortality, including age, comorbidities (hypertension, diabetes), specific treatments (antibiotics, remdesivir, favipiravir, vitamin zinc), and clinical indicators (heart rate, respiratory rate, temperature). Notably, specific symptoms (productive cough, dyspnea, delirium) and laboratory values (D-dimer, ESR) also play a critical role in predicting outcomes. This study highlights the importance of feature selection and the impact of data quantity and quality on model performance.

CONCLUSION

This study highlights the potential of ML analysis to improve the accuracy of COVID-19 mortality prediction and emphasizes the need for a comprehensive approach that considers multiple feature categories. It highlights the critical role of data quality and quantity in improving model performance and contributes to our understanding of the multifaceted factors that influence COVID-19 outcomes.

Joint modeling of longitudinal and competing risks for assessing blood oxygen saturation and its association with survival outcomes in COVID-19 patients

BACKGROUND

The objective of the present study is to evaluate the association between longitudinal and survival outcomes in the presence of competing risk events. To illustrate the application of joint modeling in clinical research, we assessed the blood oxygen saturation (SPO2) and its association with survival outcomes in coronavirus disease (COVID-19).

MATERIALS AND METHODS

In this prospective cohort study, we followed 300 COVID-19 patients, who were diagnosed with severe COVID-19 in the Rohani Hospital in Babol, the north of Iran from October 22, 2020 to March 5, 2021, where death was the event of interest, surviving was the competing risk event and SPO2 was the longitudinal outcome. Joint modeling analyses were compared to separate analyses for these data.

RESULT

The estimation of the association parameter in the joint modeling verified the association between longitudinal outcome SPO2 with survival outcome of death (Hazard Ratio (HR) = 0.33, P = 0.001) and the competing risk outcome of surviving (HR = 4.18, P < 0.001). Based on the joint modeling, longitudinal outcome (SPO2) decreased in hypertension patients (β = -0.28, P = 0.581) and increased in those with a high level of SPO2 on admission (β = 0.75, P = 0.03). Also, in the survival submodel in the joint model, the risk of death survival outcome increased in patients with diabetes comorbidity (HR = 4.38, P = 0.026).

CONCLUSION

The association between longitudinal measurements of SPO2 and survival outcomes of COVID-19 confirms that SPO2 is an important indicator in this disease. Thus, the application of this joint model can provide useful clinical evidence in the different areas of medical sciences.

COVID-19 patients hospitalized after the fourth wave of the pandemic period in Vietnam: Clinical, laboratory, therapeutic features, and clinical outcomes

BACKGROUND/PURPOSE

Despite having relatively high COVID-19 vaccine coverage in Vietnam, a fraction of COVID-19 patients required hospitalization due to severe symptoms. The purpose of this study was to describe the clinical, laboratory, complications, and treatment of COVID-19 patients hospitalized during the pandemic's fourth wave.

METHODS

Genome sequencing was performed on COVID-19 patients. Data on clinical characteristics, treatment, complications, and outcomes were consistently collected.

RESULTS

The clinical classifications were mild (37.43%), moderate (24.2%), and severe (38.37%). Patients with co-morbidities, high fever >39 °C, hypertension, tachycardia, tachypnea, and SpO2<90%, had a 1.2-4 folds higher of severe progression than those with mild/moderate. Serious consequences were much more common in the severe patients than in the mild/moderate. The respiratory system of severe patients was generally documented as fine, coarse crackles, and CT scanner shown ground glass, consolidation, and opacity, with Delta variant accounting for 92.6%. Complications were common in the severe patients, including bacteria pneumonia (36.42%), ARDS (61.11%), blood clotting disorder (7.14%), infection (46.92%), and acute kidney injury (12.35%). Antiviral, antifungal, corticosteroid, anticoagulant, and ECMO regimens were utilized. Patients died mostly as a result of co-morbidities, low SpO2, lung injury, and complications such as bacterial + fungal pneumonia (83.9%), ARDS (83.9%), bacteremia (56.5%), injury acute renal failure (27.4%), and coagulopathy (12.9%).

CONCLUSION

Severe and critical COVID-19 patients frequently have several comorbidities, multiple lung lesions along with a variety of clinical signs. Despite receiving antivirals, antibiotics, corticosteroids, anticoagulants, and even ECMO therapy, the patient encountered multiple complications, with a fatality rate of up to 38.27%.

Baseline and early changes in laboratory parameters predict disease severity and fatal outcomes in COVID-19 patients

Introduction

Coronavirus disease 2019 (COVID-19) has become the worst catastrophe of the twenty-first century and has led to the death of more than 6.9 million individuals across the globe. Despite the growing knowledge of the clinicopathological features of COVID-19, the correlation between baseline and early changes in the laboratory parameters and the clinical outcomes of patients is not entirely understood.

Methods

Here, we conducted a time series cross-sectional study aimed at assessing different measured parameters and socio-demographic factors that are associated with disease severity and the outcome of the disease in 268 PCR-confirmed COVID-19 Patients.

Results

We found COVID-19 patients who died had a median age of 61 years (IQR, 50 y - 70 y), which is significantly higher (p < 0.05) compared to those who survived and had a median age of 54 years (IQR, 42y - 65y). The median RBC count of COVID-19 survivors was 4.9 × 106/μL (IQR 4.3 × 106/μL - 5.2 × 106/μL) which is higher (p < 0.05) compared to those who died 4.4 × 106/μL (3.82 × 106/μL - 5.02 × 106/μL). Similarly, COVID-19 survivors had significantly (p < 0.05) higher lymphocyte and monocyte percentages compared to those who died. One important result we found was that COVID-19 patients who presented with severe/critical cases at the time of first admission but managed to survive had a lower percentage of neutrophil, neutrophil to lymphocyte ratio, higher lymphocyte and monocyte percentages, and RBC count compared to those who died.

Conclusion

To conclude here, we showed that simple laboratory parameters can be used to predict severity and outcome in COVID-19 patients. As these parameters are simple, inexpensive, and radially available in most resource-limited countries, they can be extrapolated to future viral epidemics or pandemics to allocate resources to particular patients.

The new combination of oxygen saturation with age shock index predicts the outcome of COVID-19 pneumonia

Introduction

Emergency departments around the world have been struggling to deal with patients with COVID-19 and presumed COVID-19. Triaging patients who need further medical support is the key matter to emergency physicians as the delay of proper treatment may worsen the results. The aim of this study was to validate the ability of age shock index and hypoxia-age-shock index at the time of presentation to the emergency department to predict case fatality in patients with COVID-19 pneumonia.

Methods

We only included patients who had COVID-19-associated pneumonia who needed in-hospital treatment. The vital signs and oxygen saturation used in the study were collected, especially from the triage sector, before patients were given supplemental oxygen.

Results

A total of 241 patients enrolled in the study. The case fatality rate was 27%. The median age of the study samples was 78 (66-86) years with 133 male and 108 female patients. Hypoxia-age-shock index showed the best performance in analysis (odds ratio 15.1, 95% confidence interval: 5.1-44.4; adjusted odds ratio 8.6, 95% confidence interval: 2.8-26.8).

Conclusion

The hypoxia-age-shock index was a strong predictor for in-hospital mortality of COVID-19 pneumonia. Furthermore, when it was compared with age shock index, hypoxia-age-shock index showed better performance in predicting fatality of the disease.

Predicting mortality among patients with severe COVID-19 pneumonia based on admission vital sign indices: a retrospective cohort study

BACKGROUND

Coronavirus disease 2019 (COVID-19) pneumonia remains a major public health concern. Vital sign indices-shock index (SI; heart rate [HR]/systolic blood pressure [SBP]), shock index age (SIA, SI × age), MinPulse (MP; maximum HR-HR), Pulse max index (PMI; HR/maximum HR), and blood pressure-age index (BPAI; SBP/age)-are better predictors of mortality in patients with trauma compared to traditional vital signs. We hypothesized that these vital sign indices may serve as predictors of mortality in patients with severe COVID-19 pneumonia. This study aimed to describe the association between vital sign indices at admission and COVID-19 pneumonia mortality and to modify the CURB-65 with the best performing vital sign index to establish a new mortality prediction tool.

METHODS

This retrospective study was conducted at a tertiary care center in southern Thailand. Adult patients diagnosed with COVID-19 pneumonia were enrolled in this study between January 2020 and July 2022. Patient demographic and clinical data on admission were collected from an electronic database. The area under the receiver operating characteristic (AUC) curve analysis was used to assess the predictive power of the resultant multivariable logistic regression model after univariate and multivariate analyses of variables with identified associations with in-hospital mortality.

RESULTS

In total, 251 patients with COVID-19 pneumonia were enrolled in this study. The in-hospital mortality rate was 27.9%. Non-survivors had significantly higher HR, respiratory rate, SIA, and PMI and lower MP and BPAI than survivors. A cutoff value of 51 for SIA (AUC, 0.663; specificity, 80%) was used to predict mortality. When SIA was introduced as a modifier for the CURB-65 score, the new score (the CURSIA score) showed a higher AUC than the Acute Physiology and Chronic Health Evaluation II and CURB-65 scores (AUCs: 0.785, 0.780, and 0.774, respectively) without statistical significance.

CONCLUSIONS

SIA and CURSIA scores were significantly associated with COVID-19 pneumonia mortality. These scores may contribute to better patient triage than traditional vital signs.

Effects of COVID-19 disease on PAI-1 antigen and haematological parameters during disease management: A prospective cross-sectional study in a regional Hospital in Ghana

BACKGROUND

Individuals with COVID-19 experience thrombotic events probably due to the associated hypofibrinolysis resulting from the upregulation of plasminogen activator inhibitor-1 (PAI-1) antigen. This study evaluated plasma PAI-1 antigen levels and haematological parameters before treatment and after recovery from severe COVID-19 in Ghana.

MATERIALS AND METHODS

This cross-sectional study was conducted at Sunyani Regional Hospital, and recruited 51 patients who had RT-PCR-confirmed SARS-CoV-2. Participants' sociodemographic data and clinical characteristics were taken from the hospital records. Venous blood was taken before COVID-19 treatment commenced for FBC, PAI-1 and ferritin assays. FBC was assessed using an automated haematology analyzer, whilst plasma PAI-1 Ag and serum ferritin levels were assessed with sandwich ELISA. All the tests were repeated immediately after participants recovered from COVID-19.

RESULTS

Of the 51 participants recruited into the study, 78.4% (40) had non-severe COVID-19 whiles 21.6% (11) experienced a severe form of the disease. Severe COVID-19 participants had significantly lower haemoglobin (g/dL): 8.1 (7.3-8.4) vs 11.8 (11.0-12.5), p<0.001; RBC x 1012/L: 2.9 (2.6-3.1) vs 3.4 (3.1-4.3), p = 0.001; HCT%: 24.8 ± 2.6 vs 35.3 ± 6.7, p<0.001 and platelet x 109/L: 86.4 (62.2-91.8) vs 165.5 (115.1-210.3), p<0.001, compared with the non-severe COVID-19 group. But WBC x 109/L: 11.6 (9.9-14.2) vs 5.4 (3.7-6.6), p<0.001 and ferritin (ng/mL): 473.1 (428.3-496.0) vs 336.2 (249.9-386.5), p<0.001, were relatively higher in the participants with severe COVID-19 than the non-severe COVID-19 counterparts. Also, the severely ill SARS-CoV-2-infected participants had relatively higher plasma PAI-1 Ag levels (ng/mL): 131.1 (128.7-131.9) vs 101.3 (92.0-116.8), p<0.001, than those with the non-severe form of the disease. Participants had lower haemoglobin (g/dL): 11.4 (8.8-12.3 vs 12.4 (11.5-13.6), p<0.001; RBC x 1012/L: 3.3 (2.9-4.0) vs 4.3 (3.4-4.6), p = 0.001; absolute granulocyte count x 109/L: 2.3 ± 1.0 vs 4.6 ± 1.8, p<0.001, and platelet x 109/L: 135.0 (107.0-193.0) vs 229.0 (166.0-270.0), p<0.001 values at admission before treatment commenced, compared to when they recovered from the disease. Additionally, the median PAI-1 Ag (ng/mL): 89.6 (74.9-100.8) vs 103.1 (93.2-128.7), p<0.001 and ferritin (ng/mL): 242.2 (197.1-302.1) vs 362.3 (273.1-399.9), p<0.001 levels were reduced after a successful recovery from COVID-19 compared to the values at admission.

CONCLUSION

Plasma PAI-1 Ag level was higher among severe COVID-19 participants. The COVID-19-associated inflammation could affect red blood cell parameters and platelets. Successful recovery from COVID-19, with reduced inflammatory response as observed in the decline of serum ferritin levels restores the haematological parameters. Plasma levels of PAI-1 should be assessed during the management of severe COVID-19 in Ghana. This will enhance the early detection of probable thrombotic events and prompts Physicians to provide interventions to prevent thrombotic complications associated with COVID-19.

Ethnic inequalities in COVID-19 infection, hospitalisation, intensive care admission, and death: a global systematic review and meta-analysis of over 200 million study participants

Background

COVID-19 has exacerbated existing ethnic inequalities in health. Little is known about whether inequalities in severe disease and deaths, observed globally among minoritised ethnic groups, relates to greater infection risk, poorer prognosis, or both. We analysed global data on COVID-19 clinical outcomes examining inequalities between people from minoritised ethnic groups compared to the ethnic majority group.

Methods

Databases (MEDLINE, EMBASE, EMCARE, CINAHL, Cochrane Library) were searched from 1st December 2019 to 3rd October 2022, for studies reporting original clinical data for COVID-19 outcomes disaggregated by ethnicity: infection, hospitalisation, intensive care unit (ICU) admission, and mortality. We assessed inequalities in incidence and prognosis using random-effects meta-analyses, with Grading of Recommendations Assessment, Development, and Evaluation (GRADE) use to assess certainty of findings. Meta-regressions explored the impact of region and time-frame (vaccine roll-out) on heterogeneity. PROSPERO: CRD42021284981.

Findings

77 studies comprising over 200,000,000 participants were included. Compared with White majority populations, we observed an increased risk of testing positive for infection for people from Black (adjusted Risk Ratio [aRR]:1.78, 95% CI:1.59-1.99, I2 = 99.1), South Asian (aRR:3.00, 95% CI:1.59-5.66, I2 = 99.1), Mixed (aRR:1.64, 95% CI:1.02-1.67, I2 = 93.2) and Other ethnic groups (aRR:1.36, 95% CI:1.01-1.82, I2 = 85.6). Black, Hispanic, and South Asian people were more likely to be seropositive. Among population-based studies, Black and Hispanic ethnic groups and Indigenous peoples had an increased risk of hospitalisation; Black, Hispanic, South Asian, East Asian and Mixed ethnic groups and Indigenous peoples had an increased risk of ICU admission. Mortality risk was increased for Hispanic, Mixed, and Indigenous groups. Smaller differences were seen for prognosis following infection. Following hospitalisation, South Asian, East Asian, Black and Mixed ethnic groups had an increased risk of ICU admission, and mortality risk was greater in Mixed ethnic groups. Certainty of evidence ranged from very low to moderate.

Interpretation

Our study suggests that systematic ethnic inequalities in COVID-19 health outcomes exist, with large differences in exposure risk and some differences in prognosis following hospitalisation. Response and recovery interventions must focus on tackling drivers of ethnic inequalities which increase exposure risk and vulnerabilities to severe disease, including structural racism and racial discrimination.

Funding

ESRC:ES/W000849/1.

Detecting Patient Deterioration Early Using Continuous Heart rate and Respiratory rate Measurements in Hospitalized COVID-19 Patients

BACKGROUND

Presenting symptoms of COVID-19 patients are unusual compared with many other illnesses. Blood pressure, heart rate, and respiratory rate may stay within acceptable ranges as the disease progresses. Consequently, intermittent monitoring does not detect deterioration as it is happening. We investigated whether continuously monitoring heart rate and respiratory rate enables earlier detection of deterioration compared with intermittent monitoring, or introduces any risks.

METHODS

When available, patients admitted to a COVID-19 ward received a wireless wearable sensor which continuously measured heart rate and respiratory rate. Two intensive care unit (ICU) physicians independently assessed sensor data, indicating when an intervention might be necessary (alarms). A third ICU physician independently extracted clinical events from the electronic medical record (EMR events). The primary outcome was the number of true alarms. Secondary outcomes included the time difference between true alarms and EMR events, interrater agreement for the alarms, and severity of EMR events that were not detected.

RESULTS

In clinical practice, 48 (EMR) events occurred. None of the 4 ICU admissions were detected with the sensor. Of the 62 sensor events, 13 were true alarms (also EMR events). Of these, two were related to rapid response team calls. The true alarms were detected 39 min (SD = 113) before EMR events, on average. Interrater agreement was 10%. Severity of the 38 non-detected events was similar to the severity of 10 detected events.

CONCLUSION

Continuously monitoring heart rate and respiratory rate does not reliably detect deterioration in COVID-19 patients when assessed by ICU physicians.

Clinical profile, treatment, and outcomes of patients with COVID-19 in a tertiary referral hospital in South Sumatera, Indonesia: A retrospective single-center study

BACKGROUND

Although there are fewer COVID-19 cases in Indonesia, the pandemic is still ongoing. COVID-19 has a significant death rate in Indonesia, but lack of information on the effect of different clinical and demographic factors on COVID-19-related grimness and mortality in Indonesia.

OBJECTIVE

This study examined the clinical profile, treatment, and outcomes of patients with COVID-19 at Lahat Regency Hospital in South Sumatera, Indonesia, to find relevant markers that might be utilized to predict the prognosis of these patients.

METHODS

This was a retrospective single-center study of all medical record files of confirmed patients with COVID-19 admitted to Lahat Hospital from September 2020 to August 2021 (n = 285). Descriptive statistics, Chi-square, Mann-Whitney, Multiple Logistic Regression, and Cox's proportional hazards model were used for data analyses.

RESULTS

This study included 65 non-hospitalized and 220 hospitalized patients. Hospitalized patients were divided into dead and alive groups. The median age was lower in the non-hospitalized group without gender discrimination, and most hospitalized patients had comorbidities. Vital signs and clinical features were significantly different in hospitalized patients compared to non-hospitalized. The survival patients in the hospitalized group showed lower white blood cell (WBC), neutrophil percentages, and neutrophil-lymphocyte ratio (NLR) but higher lymphocyte and eosinophil. Non-survival patients had elevated alanine aminotransferase (ALT), blood urea nitrogen (BUN), creatinine, blood glucose, and potassium. The use of Favipiravir and Remdesivir was significant between the alive and dead groups. The mean hospital stay for all patients was 9.49 ± 4.77 days, while the median duration of hospital time was 10.73 ± 4.33 days in the survival group and 5.39 ± 3.78 days in the non-survival group. Multiple logistic regression analysis determined respiration rate, WBC, and BUN as predictors of survival.

CONCLUSIONS

Age and comorbidities are significant elements impacting the seriousness of COVID-19. Abnormal signs in laboratory markers can be used as early warning and prognostic signs to prevent severity and death. Potential biomarkers at various degrees in patients with COVID-19 may also aid healthcare professionals in providing precision medicine and nursing.