Comparison of D-dimer Level Measured on Third Day of Hospitalization with Admission D-dimer Level in Predicting In-hospital Mortality in COVID-19 Patients

Objective: The present study aimed to compare the value of D-dimer measured on the 3rd day of hospitalization with admission D-dimer level in predicting in-hospital mortality in coronavirus disease 2019 (COVID-19) cases. Method: In total, 231 patients with COVID-19 disease were included in the study. D-dimer levels were estimated using immunoturbidimetric assay with normal range of 0-500 μg/mL. In the current research, the primary outcome was the in-hospital mortality. Results: In the present research, 39 (16.8%) COVID-19 cases died during the index hospitalization. In a multivariable analysis; age, D-dimer (3rd day) (OR: 1.00, 95% CI: 1.00-1.00, p<0.001), WBC count, and creatinine were independent predictors of the in-hospital death for COVID-19 cases. The ideal value of D-dimer level on the 3rd day of hospitalization was 774 μq/mL (area under curve (AUC): 0.903, 95% CI: 0.836-0.968; p<0.01) with sensitivity of 83.2% and specificity of 83.6%. It was noted that D-dimer level on the 3rd day of hospitalization had a higher sensitivity (83.2% vs 67.6%, respectively) and AUC value than that of D-dimer level on admission (0.903 vs 0.799, respectively). Conclusion: The main finding in this investigation was that D-dimer elevation on the 3rd of hospitalization is more sensitive predictor of in-hospital mortality than D-dimer elevation on admission in COVID-19 patients. Even though further investigations are needed to forecast precise prognosis in patients with COVID-19 disease in terms of D-dimer levels, we believe that D-dimer levels on the 3rd day of hospitalization have an enhanced potential to be used as a prognostic marker in routine clinical practice.

Download Full-text

Comparison of D-dimer level measured on the third day of hospitalization with D-dimer level in predicting in-hospital mortality in COVID-19 patients

European Heart Journal ◽

10.1093/eurheartj/ehab724.1897 ◽

2021 ◽

Vol 42 (Supplement_1) ◽

Author(s):

M Selcuk ◽

T Cinar ◽

N Gunay ◽

M Keskin ◽

V Cicek ◽

...

Keyword(s):

Hospital Mortality ◽

Primary Outcome ◽

Multivariable Analysis ◽

Hospital Death ◽

Funding Sources ◽

Auc Value ◽

Wbc Count ◽

D Dimer ◽

Higher Sensitivity ◽

The Ideal

Abstract Objective The present study aimed to compare the value of D-dimer measured on the 3rd day of hospitalization with admission D-dimer level in predicting in-hospital mortality in coronavirus disease 2019 (COVID-19) cases. Method In total, 231 patients with COVID-19 disease were included in the study. D-dimer levels were estimated using immunoturbidimetric assay with normal range of 0–500 μg/mL. In the current research, the primary outcome was the in-hospital mortality. Results In the present research, 39 (16.8%) COVID-19 cases died during the index hospitalization. In a multivariable analysis; age, D-dimer (3rd day) (OR: 1.00, 95% CI: 1.00–1.00, p<0.001), WBC count, and creatinine were independent predictors of the in-hospital death for COVID-19 cases. The ideal value of D-dimer level on the 3rd day of hospitalization was 774 μq/mL (area under curve (AUC): 0.903, 95% CI: 0.836–0.968; p<0.01) with sensitivity of 83.2% and specificity of 83.6%. It was noted that D-dimer level on the 3rd day of hospitalization had a higher sensitivity (83.2% vs 67.6%, respectively) and AUC value than that of D-dimer level on admission (0.903 vs 0.799, respectively). Conclusion The main finding in this investigation was that D-dimer elevation on the 3rd of hospitalization is more sensitive predictor of in-hospital mortality than D-dimer elevation on admission in COVID-19 patients. Even though further investigations are needed to forecast precise prognosis in patients with COVID-19 disease in terms of D-dimer levels, we believe that D-dimer levels on the 3rd day of hospitalization have an enhanced potential to be used as a prognostic marker in routine clinical practice. FUNDunding Acknowledgement Type of funding sources: None. Table 1 Figure 1

Download Full-text

265 Incidence, determinants and prognostic relevance of HS-troponin and natriuretic peptides elevation at admission in hospitalized COVID-19 pneumonia patients

European Heart Journal Supplements ◽

10.1093/eurheartj/suaa198 ◽

2020 ◽

Vol 22 (Supplement_N) ◽

pp. N65-N79

Author(s):

Luca Arcari ◽

Michelangelo Luciano ◽

Luca Cacciotti ◽

Maria Beatrice Musumeci ◽

Valerio Spuntarelli ◽

...

Keyword(s):

Hospital Mortality ◽

Natriuretic Peptides ◽

Multivariable Analysis ◽

Left Ventricular ◽

Cardiac Biomarkers ◽

Left Ventricular Ejection ◽

Ventricular Ejection Fraction ◽

Good Ability ◽

Myocardial Involvement ◽

D Dimer

Abstract Aims myocardial involvement in the course of Coronavirus disease 2019 (COVID-19) pneumonia has been reported, though not fully characterized yet. Aim of the present study is to undertake a joint evaluation of hs-Troponin and natriuretic peptides (NP) in patients hospitalized for COVID-19 pneumonia. Methods and results in this multicenter observational study, we analyzed data from n = 111 COVID-19 patients admitted to dedicated “COVID-19” medical units. Hs-Troponin was assessed in n = 103 patients and NP in n = 82 patients on admission; subgroups were identified according to values beyond reference range. increased hs-Troponin and NP were found in 38% and 56% of the cases respectively. As compared to those with normal cardiac biomarkers, these patients were older, had higher prevalence of cardiovascular diseases (CVD) and more severe COVID-19 pneumonia by higher CRP and D-dimer and lower PaO2/FIO2. Two-dimensional echocardiography performed in a subset of patients (n = 24) showed significantly reduced left ventricular ejection fraction in patients with elevated NP only (p = 0.02), whereas right ventricular systolic function (tricuspid annular plane systolic excursion) was significantly reduced both in patients with high hs-Troponin and NP (p = 0.022 and p = 0.03 respectively). On multivariable analysis, independent associations were found of hs-Troponin with age, PaO2/FIO2 and D-dimer (B = 0.419, p = 0.001; B=-0.212, p = 0.013 and B = 0.179, p = 0.037 respectively), and of NP with age and previous CVD (B = 0.480, p < 0.001 and B = 0.253, p = 0.001 respectively). In patients with in-hospital mortality (n = 23, 21%) hs-Troponin and NP were both higher (p = 0.001 and p = 0.002 respectively), while increasing hs-troponin and NP were associated with worse in-hospital prognosis [OR 4.88 (95% CI 1.9-12.2), p = 0.001 (adjusted OR 3.1 (95% CI 1.2-8.5), p = 0.025) and OR 4.67 (95% CI 2-10.8), p < 0.001 (adjusted OR 2.89 (95% CI 1.1-7.9), p = 0.04) respectively]. Receiver operator characteristic curves showed good ability of hs-Troponin and NP in predicting in-hospital mortality (AUC = 0.869 p < 0.001 and AUC = 0.810, p < 0.001 respectively). Conclusion myocardial involvement at admission is common in COVID-19 pneumonia and associated to worse prognosis, suggesting a role for cardiac biomarkers assessment in COVID-19 risk stratification. Independent associations of hs-Troponin with markers of disease severity and of NP with underlying CVD might point towards existing different mechanisms leading to their elevation in this setting.

Download Full-text

Prognostic value of D-dimer for adverse outcomes in patients with infective endocarditis: an observational study

BMC Cardiovascular Disorders ◽

10.1186/s12872-021-02078-3 ◽

2021 ◽

Vol 21 (1) ◽

Author(s):

Ying-Wen Lin ◽

Mei Jiang ◽

Xue-biao Wei ◽

Jie-leng Huang ◽

Zedazhong Su ◽

...

Keyword(s):

Regression Analysis ◽

Infective Endocarditis ◽

Adverse Events ◽

Prognostic Value ◽

Multivariable Analysis ◽

Adverse Outcomes ◽

Hospital Death ◽

Large Sample Size ◽

Cox Regression Analysis ◽

D Dimer

Abstract Background Increased D-dimer levels have been shown to correlate with adverse outcomes in various clinical conditions. However, few studies with a large sample size have been performed thus far to evaluate the prognostic value of D-dimer in patients with infective endocarditis (IE). Methods 613 patients with IE were included in the study and categorized into two groups according to the cut-off of D-dimer determined by receiver operating characteristic (ROC) curve analysis for in-hospital death: > 3.5 mg/L (n = 89) and ≤ 3.5 mg/L (n = 524). Multivariable regression analysis was used to determine the association of D-dimer with in-hospital adverse events and six-month death. Results In-hospital death (22.5% vs. 7.3%), embolism (33.7% vs 18.2%), and stroke (29.2% vs 15.8%) were significantly higher in patients with D-dimer > 3.5 mg/L than in those with D-dimer ≤ 3.5 mg/L. Multivariable analysis showed that D-dimer was an independent risk factor for in-hospital adverse events (odds ratio = 1.11, 95% CI 1.03–1.19, P = 0.005). In addition, the Kaplan–Meier curve showed that the cumulative 6-month mortality was significantly higher in patients with D-dimer > 3.5 mg/L than in those with D-dimer ≤ 3.5 mg/L (log-rank test = 39.19, P < 0.0001). Multivariable Cox regression analysis showed that D-dimer remained a significant predictor for six-month death (HR 1.11, 95% CI 1.05–1.18, P < 0.001). Conclusions D-dimer is a reliable prognostic biomarker that independently associated with in-hospital adverse events and six-month mortality in patients with IE.

Download Full-text

Predictive Value of Lactate Dehydrogenase to Albumin Ratio (LAR) in Patients With Coronavirus Disease 2019 (COVID-19)

10.21203/rs.3.rs-143788/v1 ◽

2021 ◽

Author(s):

Mingqing Liu ◽

Luping Zhang ◽

Yingli Zhang ◽

Xiaoming Xu ◽

Tantan Ma ◽

...

Keyword(s):

Lactate Dehydrogenase ◽

Hospital Mortality ◽

Disease Severity ◽

Predictive Value ◽

Hospital Death ◽

Albumin Ratio ◽

Acute Infectious Disease ◽

Critical Patients ◽

Wbc Count

Abstract BackgroundCoronavirus Disease-2019 (COVID-19) is an emerging acute infectious disease that was first discovered in Wuhan, Hubei Province, China. Since then, it has quickly spread to over one hundred cities around the world. Therefore, it is crucial to identify the risk factors of in-hospital mortality and disease severity for COVID-19 patients. MethodsWe firstly proposed a biomarker ratio, lactate dehydrogenase(LDH) to albumin ratio (LAR) may be more reliable to assess the predictive value of LAR for in-hospital mortality and early identification of critical COVID-19 patients. A retrospective study was conducted including patients (≥18 years old) with laboratory-confirmed COVID-19 infection who had been discharged or had died from 1 February to 29 February, 2020. ResultsThe study included 321 patients with COVID-19. The median age of the 321 patients was 63.0 (IQR 51.0-70.0), ranging from 19 to 95 years old and 180 (56.1%) patients were male. 142 (44.2%) patients had 1 or more coexisting comorbidity. The most common symptoms on admission were fever(289[90%]) and cough(258[80.4%]). In multivariable logistic regression, only older age (OR, 1.11; 95% CI, 1.05-1.16), WBC count (OR, 1.26; 95% CI, 1.11-1.44), lymphocyte count (OR, 0.78; 95% CI, 0.62-0.99) and LAR (OR, 1.29; 95% CI, 1.18-1.40) were found to be significantly associated with in-hospital death. ROC analysis showed that LAR had a higher AUC (0.917) and the highest specificity(84.0%) and sensitivity(84.6%). Furthermore, the results showed that LAR had a higher AUC (0.931) to differentiate critical from mild patients and had a sensitivity of 87.7% and a specificity of 82.1%. Besides, LAR had an AUC (0.861) to differentiate critical from severe patients and had a sensitivity of 86.0% and a specificity of 73.8% and the role of LAR to distinguish severe from mild patients was the worst. ConclusionsTo the best of our knowledge, this study is the first for us to explore the predictive value of LAR for in-hospital mortality and disease severity. A high LAR appears to predict higher odds of mortality and differentiate critical patients from mild or severe COVID-19 patients.

Download Full-text

Predictive determinants of overall survival among re-infected COVID-19 patients using the elastic-net regularized Cox proportional hazards model: a machine-learning algorithm

BMC Public Health ◽

10.1186/s12889-021-12383-3 ◽

2022 ◽

Vol 22 (1) ◽

Author(s):

Vahid Ebrahimi ◽

Mehrdad Sharifi ◽

Razieh Sadat Mousavi-Roknabadi ◽

Robab Sadegh ◽

Mohammad Hossein Khademian ◽

...

Keyword(s):

Machine Learning ◽

Overall Survival ◽

Mortality Rate ◽

Serum Creatinine ◽

Hospital Mortality ◽

Proportional Hazards ◽

Cox Proportional Hazards ◽

Elastic Net ◽

Hospital Death ◽

Wbc Count

Abstract Background Narrowing a large set of features to a smaller one can improve our understanding of the main risk factors for in-hospital mortality in patients with COVID-19. This study aimed to derive a parsimonious model for predicting overall survival (OS) among re-infected COVID-19 patients using machine-learning algorithms. Methods The retrospective data of 283 re-infected COVID-19 patients admitted to twenty-six medical centers (affiliated with Shiraz University of Medical Sciences) from 10 June to 26 December 2020 were reviewed and analyzed. An elastic-net regularized Cox proportional hazards (PH) regression and model approximation via backward elimination were utilized to optimize a predictive model of time to in-hospital death. The model was further reduced to its core features to maximize simplicity and generalizability. Results The empirical in-hospital mortality rate among the re-infected COVID-19 patients was 9.5%. In addition, the mortality rate among the intubated patients was 83.5%. Using the Kaplan-Meier approach, the OS (95% CI) rates for days 7, 14, and 21 were 87.5% (81.6-91.6%), 78.3% (65.0-87.0%), and 52.2% (20.3-76.7%), respectively. The elastic-net Cox PH regression retained 8 out of 35 candidate features of death. Transfer by Emergency Medical Services (EMS) (HR=3.90, 95% CI: 1.63-9.48), SpO2≤85% (HR=8.10, 95% CI: 2.97-22.00), increased serum creatinine (HR=1.85, 95% CI: 1.48-2.30), and increased white blood cells (WBC) count (HR=1.10, 95% CI: 1.03-1.15) were associated with higher in-hospital mortality rates in the re-infected COVID-19 patients. Conclusion The results of the machine-learning analysis demonstrated that transfer by EMS, profound hypoxemia (SpO2≤85%), increased serum creatinine (more than 1.6 mg/dL), and increased WBC count (more than 8.5 (×109 cells/L)) reduced the OS of the re-infected COVID-19 patients. We recommend that future machine-learning studies should further investigate these relationships and the associated factors in these patients for a better prediction of OS.

Download Full-text

Development and validation of a predictive model of in-hospital mortality in COVID-19 patients

PLoS ONE ◽

10.1371/journal.pone.0247676 ◽

2021 ◽

Vol 16 (3) ◽

pp. e0247676

Author(s):

Diego Velasco-Rodríguez ◽

Juan-Manuel Alonso-Dominguez ◽

Rosa Vidal Laso ◽

Daniel Lainez-González ◽

Aránzazu García-Raso ◽

...

Keyword(s):

Hospital Mortality ◽

Predictive Model ◽

Predictive Value ◽

Demographic Data ◽

Smoking Habit ◽

Fatal Outcome ◽

Multivariable Analysis ◽

Male Gender ◽

Correlated Factors ◽

D Dimer

We retrospectively evaluated 2879 hospitalized COVID-19 patients from four hospitals to evaluate the ability of demographic data, medical history, and on-admission laboratory parameters to predict in-hospital mortality. Association of previously published risk factors (age, gender, arterial hypertension, diabetes mellitus, smoking habit, obesity, renal failure, cardiovascular/ pulmonary diseases, serum ferritin, lymphocyte count, APTT, PT, fibrinogen, D-dimer, and platelet count) with death was tested by a multivariate logistic regression, and a predictive model was created, with further validation in an independent sample. A total of 2070 hospitalized COVID-19 patients were finally included in the multivariable analysis. Age 61–70 years (p<0.001; OR: 7.69; 95%CI: 2.93 to 20.14), age 71–80 years (p<0.001; OR: 14.99; 95%CI: 5.88 to 38.22), age >80 years (p<0.001; OR: 36.78; 95%CI: 14.42 to 93.85), male gender (p<0.001; OR: 1.84; 95%CI: 1.31 to 2.58), D-dimer levels >2 ULN (p = 0.003; OR: 1.79; 95%CI: 1.22 to 2.62), and prolonged PT (p<0.001; OR: 2.18; 95%CI: 1.49 to 3.18) were independently associated with increased in-hospital mortality. A predictive model performed with these parameters showed an AUC of 0.81 in the development cohort (n = 1270) [sensitivity of 95.83%, specificity of 41.46%, negative predictive value of 98.01%, and positive predictive value of 24.85%]. These results were then validated in an independent data sample (n = 800). Our predictive model of in-hospital mortality of COVID-19 patients has been developed, calibrated and validated. The model (MRS-COVID) included age, male gender, and on-admission coagulopathy markers as positively correlated factors with fatal outcome.

Download Full-text

Abstract P161: Drug-Eluting Stent Utilization and Post-Coronary Revascularization Procedure Mortality: Do Racial Disparities Matter?

Circulation ◽

10.1161/circ.129.suppl_1.p161 ◽

2014 ◽

Vol 129 (suppl_1) ◽

Author(s):

Alexander V Sergeev

Keyword(s):

Native American ◽

Hospital Mortality ◽

Racial Disparities ◽

Coronary Revascularization ◽

Multivariable Analysis ◽

Drug Eluting Stents ◽

Drug Eluting Stent ◽

Hospital Death ◽

Drug Eluting ◽

To Receive

Background: Besides being a major risk factor for coronary artery disease (CAD), diabetes mellitus (DM) also worsens CAD patients’ prognosis. Percutaneous coronary intervention (PCI) with stenting is an effective treatment procedure for certain categories of CAD patients with DM (CAD-DM). Newer drug-eluting stents (DES) were developed to minimize the occurrence of restenosis known to hinder PCI with older non-drug-eluting stents (non-DES). We hypothesized that disparities in DES utilization and post-procedure mortality would exist in CAD-DM patients. Methods: We conducted a retrospective cohort study of in-hospital mortality in 132,934 CAD-DM patients [mean+SD age: 65.3+11.4 years; 62.7% (83,409 of 132,934) males] after PCI with DES and non-DES in the years 2007,2009. Patient race was classified as white, black, or other Asian, Pacific Islander, Native American). Due to short length of stay, in-hospital death was defined as a binary variable (discharged alive vs. deceased). Multivariable logistic regression was used to obtain adjusted odds ratios (ORs) for in-hospital death after DES and non-DES PCI in relation to patients’ race, adjusting for confounders: age, gender, comorbidities, number of stents inserted, number of vessels treated, socio-economic status, emergency department (ED) admission, rural/urban area status. Results: Blacks were less likely to receive newer DES stent than any other racial group: 75.7% blacks, 76.9% whites, 79.0% other (p<0.04). Blacks were more likely to undergo an emergency rather than planned procedure: 45.7% blacks, 39.8% whites, and 42.7% other were ED-admitted (p<0.01). In adjusted multivariable analysis, controlling for the type of stent and confounding factors, post-PCI mortality in blacks was similar to whites (adjusted OR 0.824, 95% CI: 0.672-1.010, p=0.06), but in other was higher than in whites (adjusted OR 1.284, 95% CI: 1.151-1.434, p<0.001). Conclusions: Racial disparities exist in DES utilization by CAD-DM patients and in post-PCI in-hospital mortality: blacks are least likely to receive DES, but patients of other racial groups are more likely to die after PCI. Further studies investigating the mechanisms responsible for these disparities are warranted.

Download Full-text

Endothelium Activation Markers in Severe Hospitalized COVID-19 Patients: Role in Mortality Risk Prediction

TH Open ◽

10.1055/s-0041-1731711 ◽

2021 ◽

Vol 05 (03) ◽

pp. e253-e263

Author(s):

Marina Marchetti ◽

Patricia Gomez-Rosas ◽

Eleonora Sanga ◽

Sara Gamba ◽

Cristina Verzeroli ◽

...

Keyword(s):

Hospital Mortality ◽

Multivariable Analysis ◽

Risk Groups ◽

Endothelial Activation ◽

Endothelial Damage ◽

High Risk Group ◽

Activation Markers ◽

Soluble Thrombomodulin ◽

Inflammatory Status ◽

D Dimer

Abstract Introduction Endothelial damage and hypercoagulability are major players behind the hemostatic derangement of SARS-CoV-2 infection. Aim In this prospective study we assessed endothelial and inflammatory biomarkers in a cohort of COVID-19 patients, aiming to identify predictive factors of in-hospital mortality. Methods COVID-19 patients hospitalized in intensive care (ICU) and non-ICU units at 2 Bergamo (Italy) hospitals from March 23 to May 30, 2020, were enrolled. Markers of endothelium activation including von-Willebrand factor (vWF), soluble thrombomodulin (sTM), and fibrinolytic proteins (t-PA and PAI-1) were measured. Additionally, D-dimer, Fibrinogen, FVIII, nucleosomes, C reactive protein (CRP) and procalcitonin were assessed. Results Sixty-three (45 ICU, and 18 non-ICU) patients, with a median age of 62 years were analyzed. Increased plasma levels of D-dimer, FVIII, fibrinogen, nucleosomes, CRP, and procalcitonin were observed in the whole cohort. Extremely elevated vWF levels characterized all patients (highest values in ICU-subjects). After a median time of 30 days, death occurred in 13 (21%) patients. By multivariable analysis, vWF-activity, neutrophil-count and PaO2/FiO2 were significantly associated with death. Using these variables, a linear score with 3-risk groups was generated that provided a cumulative incidence of death of 0% in the low-, 32% in the intermediate-, and 78% in the high-risk group. Conclusions COVID-19-induced hemostatic abnormalities are exacerbated by the severity of the disease and strongly correlate with the inflammatory status, underlying the link between coagulation, endothelial activation, and inflammation. Our study provides evidence for a role of vWF, together with neutrophils and PaO2/FiO2, as a significant predictor of in-hospital mortality by SARSCoV-2 infection.

Download Full-text

“Urban Tertiary Care Centre Experience of Characteristics of Severe COVID-19 Pneumonia”

10.1101/2021.08.16.21262073 ◽

2021 ◽

Author(s):

Nehal M Shah ◽

Janakkumar R Khambholja ◽

Nilay Suthar ◽

Hemant Tiwari ◽

Vandit Desai ◽

...

Keyword(s):

Heart Rate ◽

Hospital Mortality ◽

Respiratory Rate ◽

Inflammatory Markers ◽

Tertiary Care ◽

Current Evidence ◽

Hospital Death ◽

Tertiary Care Centre ◽

Novel Coronavirus ◽

D Dimer

AbstractIntroductionThe global pandemic of novel coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) began in Wuhan, China, in December 2019, and has since spread worldwide.[1] This study attempts to summarize current evidence regarding major inflammatory markers, severity predictors and its impact on outcome, which provide current clinical experience and treatment guidance for this novel coronavirus.MethodsThis is a retrospective observational study done at an urban teaching covid-19 designated hospital. Hospital data were analysed with aim of studying inflammatory markers, predictors and outcome. Patients were classified in Mild, Moderate, Severe & Critical categories of COVID cases. Their clinical parameters, laboratory investigations, radiological findings & Outcome measures were studied. Strength of association & correlation of those parameters with severity and in-hospital mortality were studied.ResultsA total 204 (N) patients were clinically classified into different severity groups, as per MOHFW and qCSI(quick Covid Severity Index) guidelines, as Mild (34), Moderate (56), Severe (39) and Critical (75). The mean(SD) age of the cohort was 55.1+13.2 years; 74.02% were male. Severe COVID-19 illness is seen more in patients more than 50 years of age. COVID-19 patients having IHD develop worse disease with excess early in-hospital mortality. Respiratory rate & Heart Rate on admission are correlated with severe and stormy disease. Among Inflammatory markers, on admission LDH, D-Dimer and CRP are related with severity and excess in-hospital death rate.ConclusionAdvanced age, male gender, IHD, Respiratory Rate & Heart Rate on admission were associated with severe covid-19 illness. S. Lactate Dehydrogenase & D-dimer was associated with severe covid-19 illness and early in-hospital death.

Download Full-text

Predictive determinants of overall survival among re-infected COVID-19 patients using the elastic-net regularized Cox proportional hazards model: a machine-learning algorithm

10.21203/rs.3.rs-371941/v1 ◽

2021 ◽

Author(s):

Vahid Ebrahimi ◽

Mehrdad Sharifi ◽

Razieh Sadat Mousavi-Roknabadi ◽

Robab Sadegh ◽

Mohammad Hossein Khademian ◽

...

Keyword(s):

Machine Learning ◽

Overall Survival ◽

Mortality Rate ◽

Serum Creatinine ◽

Hospital Mortality ◽

Proportional Hazards ◽

Cox Proportional Hazards ◽

Elastic Net ◽

Hospital Death ◽

Wbc Count

Abstract Background: Narrowing a large set of features to a smaller one can improve our understanding of the main risk factors for in-hospital mortality in patients with COVID-19. This study aimed to derive a parsimonious model for predicting overall survival (OS) among re-infected COVID-19 patients using machine-learning (ML) algorithms.Methods: The retrospective data of 283 re-infected COVID-19 patients admitted to twenty-six medical centers (affiliated with Shiraz University of Medical Sciences) from 10 June to 26 December 2020 were reviewed and analyzed. An elastic-net regularized Cox proportional hazards (PH) regression and model approximation via backward elimination were utilized to optimize a predictive model of time to in-hospital death. Then, the model was reduced to its core features to maximize simplicity and generalizability.Results: The empirical in-hospital mortality rate among the re-infected COVID-19 patients was obtained as 9.5%. In addition, the mortality rate among the intubated patients was 83.5%. Using the Kaplan-Meier approach, the OS (95% CI) rates for days 7, 14, and 21 were 87.5% (81.6-91.6%), 78.3% (65.0-87.0%), and 52.2% (20.3-76.7%), respectively. The elastic-net Cox PH regression retained 8 out of 35 candidate features of death. Transfer by Emergency Medical Services (EMS) (HR=3.90, 95% CI: 1.63-9.48), SpO2≤85% (HR=8.10, 95% CI: 2.97-22.00), increased serum creatinine (HR=1.85, 95% CI: 1.48-2.30), and increased white blood cells (WBC) count (HR=1.10, 95% CI: 1.03-1.15) were associated with higher in-hospital mortality rates in the re-infected COVID-19 patients.Conclusion: The results of the ML analysis demonstrated that transfer by EMS, profound hypoxemia (SpO2≤85%), increased serum creatinine (more than 1.6 mg/dL), and increased WBC count (more than 8.5 (×109 cells/L)) reduced the OS of the re-infected COVID-19 patients.

Download Full-text