INTRO    EQUATIONS    CRITERIA    DECISION    CONVERT    SPECIALTIES    REFS    SEARCH    NEW    FHIR APP    POLICIES    CONTACT

 

COVID-19 Criticality Prediction (Yan 3/3/2020)

 

Input
 

LDH		IU/L IntUnit/L U/L Unit/L Units/L mcKat/L
hsCRP		mg/L mg/L mg/dL nmol/L
Percent Lymphocytes		% fraction percent rate ratio

 
Results
 
Important: Inputs must be complete to perform calculation.

 

Prediction

 

Data action

 
     
 
Include literature references: Yes   No

 

 

Notes

• This calculator represents findings from the literature referenced below. It is important to understand that these findings may not be representative of the experience in your local healthcare system.
• This algorithm is based on a pre-print publication^[1] and has not completed peer-review or methodological review yet. The results cannot be verified to be clinically accurate at your organization.
• LDH: lactate dehydrogenase; hsCRP: high sensitivity C-reactive protein.

 

Calculation Details and Variables

• Three predictive factors were discovered through machine learning using the XGBoost Classifier.^[2]
• A decision tree analysis starting with LDH measurement, and later hsCRP and the Percent Lmphocytes values were found to predict survival.
• LDH >=365 U/L was associated with increased risk of death.
• For lower LDH levels, an hsCRP level <41.2 mg/L was associated with increased probability of survival.
• If the LDH was <365 U/L and the hsCRP was ≥41.2 mg/L, the Percent Lymphocyte value predicted outcome: In these cases, Percent Lymphocyte >14.7% were associated with increased survival while Percent Lymphocyte ≤14.7% were associated with increased mortality.

 

References

1. Yan L, Zhang HT, Xiao Y, et al. Prediction of criticality in patients with severe Covid-19 infection using three clinical features: a machine learning-based prognostic model with clinical data in Wuhan. MedRxiv, the Preprint Server for health Sciences. https://www.medrxiv.org/content/10.1101/2020.02.27.20028027v2 accessed 3/17/2020.
2. Chen T, Guestrin C. Xgboost: A scalable tree boosting system. arxiv e-prints. arXiv preprint arXiv:1603.0275, 2016.

 

Associated Logical Observation Identifiers Names and Codes [LOINC]
 

Parameter	LOINC's and Suggested Units
LDH	14804-9 : U/L      14805-6 : U/L      2537-9 : U/L
hsCRP	1988-5 : mg/L      30522-7 : mg/L      71426-1 : mg/L      76486-0 : nmol/L
Percent Lymphocytes	736-9 : %      735-1 : %      737-7 : %      15197-7 : %      26478-8 : %

 
This material contains content from LOINC (http://loinc.org). LOINC is copyright © 1995-2023, Regenstrief Institute, Inc. and the Logical Observation Identifiers Names and Codes (LOINC) Committee and is available at no cost under the license at http://loinc.org/license. LOINC® is a registered United States trademark of Regenstrief Institute, Inc.