Skip to main content

Table 3 Summary of severity prediction efficiency of COVID-19 patients using Model-A, LMN index and other markers in Hefei cohort

From: Lymphocyte–monocyte–neutrophil index: a predictor of severity of coronavirus disease 2019 patients produced by sparse principal component analysis

 

Cutoff

AUC(95% CI)

ACC(95% CI)

SEN(95% CI)

SPE(95% CI)

PPV(95% CI)

NPV(95% CI)

Model-A

0.125

0.867 (0.786–0.947)

0.726 (0.721–0.731)

1 (1–1)

0.574 (0.433–0.716)

0.565 (0.422–0.708)

1 (1–1)

LMN

0.281

0.837 (0.75–0.925)

0.793 (0.789–0.797)

0.786 (0.634–0.938)

0.796 (0.689–0.904)

0.667 (0.506–0.828)

0.878 (0.786–0.969)

ALB

42.05

0.846 (0.758–0.934)

0.79 (0.786–0.794)

0.821 (0.68–0.963)

0.774 (0.661–0.886)

0.657 (0.5–0.814)

0.891 (0.801–0.981)

CD4

352.5

0.773 (0.656–0.89)

0.721 (0.715–0.728)

0.75 (0.577–0.923)

0.703 (0.555–0.85)

0.621 (0.444–0.797)

0.812 (0.677–0.948)

CD8

146.5

0.809 (0.701–0.917)

0.77 (0.765–0.776)

0.625 (0.431–0.819)

0.865 (0.755–0.975)

0.75 (0.56–0.94)

0.78 (0.654–0.907)

CRP

7.6

0.811 (0.716–0.906)

0.683 (0.678–0.688)

0.964 (0.896–1.033)

0.537 (0.404–0.67)

0.519 (0.383–0.655)

0.967 (0.902–1.031)

DD

0.44

0.754 (0.641–0.868)

0.756 (0.752–0.761)

0.536 (0.351–0.72)

0.88 (0.79–0.97)

0.714 (0.521–0.908)

0.772 (0.663–0.881)

IL6

7.005

0.8 (0.691–0.909)

0.805 (0.801–0.809)

0.607 (0.426–0.788)

0.907 (0.83–0.985)

0.773 (0.598–0.948)

0.817 (0.719–0.915)

Lym

1.235

0.75 (0.641–0.86)

0.646 (0.641–0.652)

0.857 (0.728–0.987)

0.537 (0.404–0.67)

0.49 (0.35–0.63)

0.879 (0.767–0.99)

Lym%

21

0.833 (0.742–0.924)

0.793 (0.789–0.797)

0.857 (0.728–0.987)

0.759 (0.645–0.873)

0.649 (0.495–0.802)

0.911 (0.828–0.994)

Neu

3.825

0.786 (0.675–0.898)

0.768 (0.764–0.773)

0.786 (0.634–0.938)

0.759 (0.645–0.873)

0.629 (0.468–0.789)

0.872 (0.777–0.968)

Neu%

69.75

0.833 (0.745–0.922)

0.756 (0.752–0.761)

0.857 (0.728–0.987)

0.704 (0.582–0.825)

0.6 (0.448–0.752)

0.905 (0.816–0.994)

NLR

3.531

0.851 (0.765–0.937)

0.817 (0.814–0.821)

0.857 (0.728–0.987)

0.796 (0.689–0.904)

0.686 (0.532–0.84)

0.915 (0.835–0.995)

  1. Disease severity prediction models: Model-A and LMN index were produced by SPCA and logistic regression from Hefei Cohort. Prediction efficiency of these models and clinical markers were assessed using ROC and AUC. ACC, SEN, SPE, PPV, NPV were calculated
  2. ALB albumin, AUC area under curve, CD4 CD4 + T lymphocytes, CD8 CD8 + T lymphocytes, CI confidence interval, COVID-19 coronavirus disease 2019, CRP C-reactive protein, DD D-Dimer, IL6 interleukin-6, LMN index lymphocyte–monocyte–neutrophil index, simplified version of Model-A, LYM lymphocyte count, LYM% lymphocyte percentage, Model-A prediction model based on the first and 12th principal components produced by sparse principal component analysis, NEU neutrophil count, NEU%, eutrophils percentage, NLR netrophil-to-lymphocyte ratio, NPV negative predictive value, PPV positive predictive value, NPV negative predictive value, PPV positive predictive value
Back to article page

Virology Journal

ISSN: 1743-422X