Deep learning prediction of likelihood of ICU admission and mortality in COVID-19 patients using clinical variables

Background This study aimed to develop a deep-learning model and a risk-score system using clinical variables to predict intensive care unit (ICU) admission and in-hospital mortality in COVID-19 patients. Methods This retrospective study consisted of 5,766 persons-under-investigation for COVID-19 between 7 February 2020 and 4 May 2020. Demographics, chronic comorbidities, vital signs, symptoms and laboratory tests at admission were collected. A deep neural network model and a risk-score system were constructed to predict ICU admission and in-hospital mortality. Prediction performance used the receiver operating characteristic area under the curve (AUC). Results The top ICU predictors were procalcitonin, lactate dehydrogenase, C-reactive protein, ferritin and oxygen saturation. The top mortality predictors were age, lactate dehydrogenase, procalcitonin, cardiac troponin, C-reactive protein and oxygen saturation. Age and troponin were unique top predictors for mortality but not ICU admission. The deep-learning model predicted ICU admission and mortality with an AUC of 0.780 (95% CI [0.760–0.785]) and 0.844 (95% CI [0.839–0.848]), respectively. The corresponding risk scores yielded an AUC of 0.728 (95% CI [0.726–0.729]) and 0.848 (95% CI [0.847–0.849]), respectively. Conclusions Deep learning and the resultant risk score have the potential to provide frontline physicians with quantitative tools to stratify patients more effectively in time-sensitive and resource-constrained circumstances.

Download Full-text

446. Prognostic Value of Absolute Lymphocyte Count for Disease Severity and Clinical Outcomes in Adult COVID-19 Inpatients

Open Forum Infectious Diseases ◽

10.1093/ofid/ofab466.645 ◽

2021 ◽

Vol 8 (Supplement_1) ◽

pp. S324-S324

Author(s):

Jianli Niu ◽

Candice Sareli ◽

Maria Deane ◽

Aharon E Sareli

Keyword(s):

Lactate Dehydrogenase ◽

Clinical Outcomes ◽

Disease Severity ◽

Interleukin 6 ◽

Lymphocyte Count ◽

Absolute Lymphocyte Count ◽

Serum Levels ◽

C Reactive Protein ◽

Icu Admission ◽

Reactive Protein

Abstract Background Lymphopenia has been reported as a relatively frequent finding in patients with coronavirus disease 2019 (COVID-19). This study aimed to assess the use of absolute lymphocyte count (ALC) as a prognostic biomarker for disease severity and clinical outcomes. Methods A cohort of adult patients with COVID-19 admitted to Memorial Healthcare System, Hollywood, Florida from March 7, 2020 to January 18, 2021 was retrospectively analyzed. An absolute lymphocyte count (ALC) < 1.1 × 109/L was used as cutoff point to define lymphopenia. Correlations of ALC upon admission with age and serum levels of C-reactive protein, interleukin-6, lactate dehydrogenase, and creatinine were analyzed. Univariate and multivariate regression models were developed to assess the association of lymphopenia with the risk of ICU admission and clinical outcomes. Results 4,485 hospitalized patients were included in the final analyses. Median age was 61 (interquartile range, 47-73) years and 2,311 (51.5%) were men. Lymphopenia was more frequent in patients admitted to the ICU compared to those that were not admitted to the ICU, with an odds ratio of 2.14 (95% conﬁdence interval [CI], 1.78-2.56, p < .0001) (Figure 1). The actual value of the ALC was negatively correlated with age and serum levels of C-reactive protein, interleukin-6, lactate dehydrogenase, and creatinine (all p < 0.005). Patients with lymphopenia (n=2,409) compared to those without lymphopenia (n=2,076) had multivariable-adjusted odds ratios of 1.85 (95% CI, 1.53-2.24) for ICU admission, 2.08 (95% CI, 1.67-2.58) for intubation, 1.98 (95% CI, 1.31-3.00) for development of acute kidney failure, and 2.23 (95% CI, 1.79-2.79) for in-hospital mortality (Table 1). Analyses were adjusted for age, gender, race, hypertension, diabetes, chronic obstructive pulmonary disease, chronic kidney disease, coronary artery disease, malignancy, obesity, and smoking. Conclusion Lymphopenia in adult COVID -19 hospitalized patients was associated with increased risk of disease severity (as evidenced by need for ICU admission) and poor clinical outcomes. Absolute lymphocyte count may help with prognostication in individuals hospitalized with COVID-19. Disclosures All Authors: No reported disclosures

Download Full-text

Comparative Survival Analysis of Immunomodulatory Therapy for COVID-19 ‘Cytokine Storm’: A Retrospective Observational Cohort Study

10.1101/2020.06.16.20126714 ◽

2020 ◽

Cited By ~ 1

Author(s):

Sonali Narain ◽

Dimitre G. Stefanov ◽

Alice S. Chau ◽

Andrew G. Weber ◽

Galina Marder ◽

...

Keyword(s):

Lactate Dehydrogenase ◽

Hospital Mortality ◽

Standard Of Care ◽

P Value ◽

Cytokine Storm ◽

C Reactive Protein ◽

Reactive Protein ◽

Treatment Standard ◽

Observational Cohort ◽

Retrospective Observational Cohort Study

AbstractBackgroundCytokine storm is a marker of COVID-19 illness severity and increased mortality. Immunomodulatory treatments have been repurposed to improve mortality outcomes.MethodsWe conducted a retrospective analysis of electronic health records across the Northwell Health system. COVID-19 patients hospitalized between March 1, 2020 and April 15, 2020, were included. Cytokine storm was defined by inflammatory markers: ferritin >700ng/mL, C-reactive protein >30mg/dL, or lactate dehydrogenase >300U/L. Patients were subdivided into six groups -no immunomodulatory treatment (standard of care) and five groups that received either corticosteroids, anti-interleukin 6 (IL-6) antibody (tocilizumab) or anti-IL-1 therapy (anakinra) alone or in combination with corticosteroids. The primary outcome was hospital mortality.ResultsThere were 3,098 patients who met inclusion criteria. The most common comorbidities were hypertension (40-56%), diabetes (32-43%) and cardiovascular disease (2-15%). Patients most frequently met criteria with high lactate dehydrogenase (74.8%) alone, or in combination, followed by ferritin (71.4%) and C-reactive protein (9.4%). More than 80% of patients had an elevated D-dimer. Patients treated with a combination of tocilizumab and corticosteroids (Hazard Ratio [HR]: 0.459, 95% Confidence Interval [CI]: 0.295-0.714; p<0.0001) or corticosteroids alone (HR: 0.696, 95% CI: 0.512-0.946; p=0.01) had improved hospital survival compared to standard of care. Corticosteroids and tocilizumab was associated with increased survival when compared to corticosteroids and anakinra (HR: 0.612, 95% CI: 0.391-0.958; p-value=0.02).ConclusionsWhen compared to standard of care, corticosteroid and tocilizumab used in combination, or corticosteroids alone, was associated with reduced hospital mortality for patients with COVID-19 cytokine storm.

Download Full-text

Association of the Hospitalization and in-hospital Mortality Rates with Aspartate Aminotransferase, Alanine Aminotransferase, Lactate Dehydrogenase, C-Reactive Protein, and Prothrombin Time Levels in Cancer Patients Admitted to the Emergency Department

Turkish Journal of Oncology ◽

10.5505/tjo.2019.2252 ◽

2020 ◽

Author(s):

Yahya Ayhan ACAR

Keyword(s):

Emergency Department ◽

Lactate Dehydrogenase ◽

Hospital Mortality ◽

Cancer Patients ◽

Aspartate Aminotransferase ◽

Alanine Aminotransferase ◽

Mortality Rates ◽

C Reactive Protein ◽

Reactive Protein ◽

Lactate Dehydrogenase C

Download Full-text

Derivation and validation of a simple inflammation-based risk score system for predicting in-hospital mortality in acute coronary syndrome patients

Journal of Cardiology ◽

10.1016/j.jjcc.2018.11.010 ◽

2019 ◽

Vol 73 (5) ◽

pp. 416-424 ◽

Cited By ~ 3

Author(s):

Héctor González-Pacheco ◽

Rafael Bojalil ◽

Luis M. Amezcua-Guerra ◽

Julio Sandoval ◽

Guering Eid-Lidt ◽

...

Keyword(s):

Acute Coronary Syndrome ◽

Hospital Mortality ◽

Risk Score ◽

Score System ◽

Coronary Syndrome ◽

Risk Score System

Download Full-text

ABC2-SPH risk score for in-hospital mortality in COVID-19 patients: development, external validation and comparison with other available scores

10.1101/2021.02.01.21250306 ◽

2021 ◽

Author(s):

Milena S. Marcolino ◽

Magda C. Pires ◽

Lucas Emanuel F. Ramos ◽

Rafael T. Silva ◽

Luana M. Oliveira ◽

...

Keyword(s):

Heart Rate ◽

Hospital Mortality ◽

Blood Urea Nitrogen ◽

Risk Score ◽

Mortality Risk ◽

Scoring System ◽

List Type ◽

C Reactive Protein ◽

Urea Nitrogen ◽

Reactive Protein

AbstractObjectiveTo develop and validate a rapid scoring system at hospital admission for predicting in-hospital mortality in patients hospitalized with coronavirus disease 19 (COVID-19), and to compare this score with other existing ones.DesignCohort studySettingThe Brazilian COVID-19 Registry has been conducted in 36 Brazilian hospitals in 17 cities. Logistic regression analysis was performed to develop a prediction model for in-hospital mortality, based on the 3978 patients that were admitted between March-July, 2020. The model was then validated in the 1054 patients admitted during August-September, as well as in an external cohort of 474 Spanish patients.ParticipantsConsecutive symptomatic patients (≥18 years old) with laboratory confirmed COVID-19 admitted to participating hospitals. Patients who were transferred between hospitals and in whom admission data from the first hospital or the last hospital were not available were excluded, as well those who were admitted for other reasons and developed COVID-19 symptoms during their stay.Main outcome measuresIn-hospital mortalityResultsMedian (25th-75th percentile) age of the model-derivation cohort was 60 (48-72) years, 53.8% were men, in-hospital mortality was 20.3%. The validation cohorts had similar age distribution and in-hospital mortality. From 20 potential predictors, seven significant variables were included in the in-hospital mortality risk score: age, blood urea nitrogen, number of comorbidities, C-reactive protein, SpO2/FiO2 ratio, platelet count and heart rate. The model had high discriminatory value (AUROC 0.844, 95% CI 0.829 to 0.859), which was confirmed in the Brazilian (0.859) and Spanish (0.899) validation cohorts. Our ABC2-SPH score showed good calibration in both Brazilian cohorts, but, in the Spanish cohort, mortality was somewhat underestimated in patients with very high (>25%) risk. The ABC2-SPH score is implemented in a freely available online risk calculator (https://abc2sph.com/).ConclusionsWe designed and validated an easy-to-use rapid scoring system based on characteristics of COVID-19 patients commonly available at hospital presentation, for early stratification for in-hospital mortality risk of patients with COVID-19.Summary boxesWhat is already known on this topic?Rapid scoring systems may be very useful for fast and effective assessment of COVID-19 patients in the emergency department.The majority of available scores have high risk of bias and lack benefit to clinical decision making.Derivation and validation studies in low- and middle-income countries, including Latin America, are scarce.What this study addsABC2-SPH employs seven well defined variables, routinely assessed upon hospital presentation: age, number of comorbidities, blood urea nitrogen, C reactive protein, Spo2/FiO2 ratio, platelets and heart rate.This easy-to-use risk score identified four categories at increasing risk of death with a high level of accuracy, and displayed better discrimination ability than other existing scores.A free web-based calculator is available and may help healthcare practitioners to estimate the expected risk of mortality for patients at hospital presentation.

Download Full-text

Predictors of intubation and mortality in COVID-19 patients: a retrospective study

Journal of Anesthesia, Analgesia and Critical Care ◽

10.1186/s44158-021-00016-5 ◽

2021 ◽

Vol 1 (1) ◽

Author(s):

Tiziana Cena ◽

Gianmaria Cammarota ◽

Danila Azzolina ◽

Michela Barini ◽

Simona Bazzano ◽

...

Keyword(s):

Lactate Dehydrogenase ◽

Hospital Mortality ◽

Invasive Mechanical Ventilation ◽

Ground Glass Opacity ◽

Lactate Dehydrogenase Level ◽

University Hospital ◽

Arterial Oxygen ◽

C Reactive Protein ◽

Reactive Protein ◽

Hypoxemic Acute Respiratory Failure

Abstract Background Estimating the risk of intubation and mortality among COVID-19 patients can help clinicians triage these patients and allocate resources more efficiently. Thus, here we sought to identify the risk factors associated with intubation and intra-hospital mortality in a cohort of COVID-19 patients hospitalized due to hypoxemic acute respiratory failure (ARF). Results We included retrospectively a total of 187 patients admitted to the subintensive and intensive care units of the University Hospital “Maggiore della Carità” of Novara between March 1st and April 30th, 2020. Based on these patients’ demographic characteristics, early clinical and laboratory variables, and quantitative chest computerized tomography (CT) findings, we developed two random forest (RF) models able to predict intubation and intra-hospital mortality. Variables independently associated with intubation were C-reactive protein (p < 0.001), lactate dehydrogenase level (p = 0.018) and white blood cell count (p = 0.026), while variables independently associated with mortality were age (p < 0.001), other cardiovascular diseases (p = 0.029), C-reactive protein (p = 0.002), lactate dehydrogenase level (p = 0.018), and invasive mechanical ventilation (p = 0.001). On quantitative chest CT analysis, ground glass opacity, consolidation, and fibrosis resulted significantly associated with patient intubation and mortality. The major predictors for both models were the ratio between partial pressure of arterial oxygen and fraction of inspired oxygen, age, lactate dehydrogenase, C-reactive protein, glycemia, CT quantitative parameters, lymphocyte count, and symptom onset. Conclusions Altogether, our findings confirm previously reported demographic, clinical, hemato-chemical, and radiologic predictors of adverse outcome among COVID-19-associated hypoxemic ARF patients. The two newly developed RF models herein described show an overall good level of accuracy in predicting intra-hospital mortality and intubation in our study population. Thus, their future development and implementation may help not only identify patients at higher risk of deterioration more effectively but also rebalance the disproportion between resources and demand.

Download Full-text

Vitamin D and COVID-19 severity and related mortality: a prospective study in Italy

BMC Infectious Diseases ◽

10.1186/s12879-021-06281-7 ◽

2021 ◽

Vol 21 (1) ◽

Author(s):

Irene Campi ◽

Luigi Gennari ◽

Daniela Merlotti ◽

Christian Mingiano ◽

Alessandro Frosali ◽

...

Keyword(s):

Vitamin D ◽

Lactate Dehydrogenase ◽

C Reactive Protein ◽

Icu Admission ◽

Hydroxyvitamin D ◽

Reactive Protein ◽

Respiratory Outcome ◽

25 Hydroxyvitamin D ◽

A Prospective Study ◽

Related Mortality

Abstract Background Vitamin D deficiency has been suggested to favor a poorer outcome of Coronavirus disease-19 (COVID-19). We aimed to assess if 25-hydroxyvitamin-D (25OHD) levels are associated with interleukin 6 (IL-6) levels and with disease severity and mortality in COVID-19. Methods We prospectively studied 103 in-patients admitted to a Northern-Italian hospital (age 66.1 ± 14.1 years, 70 males) for severely-symptomatic COVID-19. Fifty-two subjects with SARS-CoV-2 infection but mild COVID-19 symptoms (mildly-symptomatic COVID-19 patients) and 206 subjects without SARS-CoV-2 infection were controls. We measured 25OHD and IL-6 levels at admission and focused on respiratory outcome during hospitalization. Results Severely-symptomatic COVID-19 patients had lower 25OHD levels (18.2 ± 11.4 ng/mL) than mildly-symptomatic COVID-19 patients and non-SARS-CoV-2-infected controls (30.3 ± 8.5 ng/mL and 25.4 ± 9.4 ng/mL, respectively, p < 0.0001 for both comparisons). 25OHD and IL-6 levels were respectively lower and higher in severely-symptomatic COVID-19 patients admitted to intensive care Unit [(ICU), 14.4 ± 8.6 ng/mL and 43.0 (19.0–56.0) pg/mL, respectively], than in those not requiring ICU admission [22.4 ± 1.4 ng/mL, p = 0.0001 and 16.0 (8.0–32.0) pg/mL, p = 0.0002, respectively]. Similar differences were found when comparing COVID-19 patients who died in hospital [13.2 ± 6.4 ng/mL and 45.0 (28.0–99.0) pg/mL] with survivors [19.3 ± 12.0 ng/mL, p = 0.035 and 21.0 (10.5–45.9) pg/mL, p = 0.018, respectively). 25OHD levels inversely correlated with: i) IL-6 levels (ρ − 0.284, p = 0.004); ii) the subsequent need of the ICU admission [relative risk, RR 0.99, 95% confidence interval (95%CI) 0.98–1.00, p = 0.011] regardless of age, gender, presence of at least 1 comorbidity among obesity, diabetes, arterial hypertension, creatinine, IL-6 and lactate dehydrogenase levels, neutrophil cells, lymphocytes and platelets count; iii) mortality (RR 0.97, 95%CI, 0.95–0.99, p = 0.011) regardless of age, gender, presence of diabetes, IL-6 and C-reactive protein and lactate dehydrogenase levels, neutrophil cells, lymphocytes and platelets count. Conclusion In our COVID-19 patients, low 25OHD levels were inversely correlated with high IL-6 levels and were independent predictors of COVID-19 severity and mortality.

Download Full-text