RESUMEN
PURPOSE: We aimed to investigate the diagnostic performance of chest CT compared with first RT-PCR results in adult patients suspected of COVID-19 infection in an ED setting. We also constructed a predictive machine learning model based on chest CT and additional data to improve the diagnostic accuracy of chest CT. METHODS: This study's cohort consisted of 319 patients who underwent chest CT and RT-PCR testing at the ED. Patient characteristics, demographics, symptoms, vital signs, laboratory tests, and chest CT results (CO-RADS) were collected. With first RT-PCR as reference standard, the diagnostic performance of chest CT using the CO-RADS score was assessed. Additionally, a predictive machine learning model was constructed using logistic regression. RESULTS: Chest CT, with first RT-PCR as a reference, had a sensitivity, specificity, PPV, and NPV of 90.2%, 88.2%, 84.5%, and 92.7%, respectively. The prediction model with CO-RADS, ferritin, leucocyte count, CK, days of complaints, and diarrhea as predictors had a sensitivity, specificity, PPV, and NPV of 89.3%, 93.4%, 90.8%, and 92.3%, respectively. CONCLUSION: Chest CT, using the CO-RADS scoring system, is a sensitive and specific method that can aid in the diagnosis of COVID-19, especially if RT-PCR tests are scarce during an outbreak. Combining a predictive machine learning model could further improve the accuracy of diagnostic chest CT for COVID-19. Further candidate predictors should be analyzed to improve our model. However, RT-PCR should remain the primary standard of testing as up to 9% of RT-PCR positive patients are not diagnosed by chest CT or our machine learning model.
Asunto(s)
Infecciones por Coronavirus/diagnóstico por imagen , Servicio de Urgencia en Hospital , Neumonía Viral/diagnóstico por imagen , Radiografía Torácica/métodos , Tomografía Computarizada por Rayos X/métodos , Triaje , Adolescente , Adulto , Anciano , Anciano de 80 o más Años , Betacoronavirus , COVID-19 , Prueba de COVID-19 , Vacunas contra la COVID-19 , Técnicas de Laboratorio Clínico , Infecciones por Coronavirus/diagnóstico , Infecciones por Coronavirus/epidemiología , Femenino , Humanos , Aprendizaje Automático , Masculino , Persona de Mediana Edad , Países Bajos/epidemiología , Pandemias , Neumonía Viral/epidemiología , Estudios Prospectivos , SARS-CoV-2 , Sensibilidad y EspecificidadRESUMEN
INTRODUCTION: The Thomas-plot has proven to be a helpful tool to discriminate between different types of anemia. This plot combines the reticulocyte hemoglobin content (Ret-He) with the soluble transferrin receptor (sTfR)/log ferritin (fer) ratio. In this study, we designed an alternative Thomas-plot in which Ret-He is combined with the transferrin (Tf)/log ferritin ratio. We validated both Thomas-plots in a population of anemic patients and compared the performance to the current laboratory diagnostics of anemia. METHODS: A total of 536 anemic patients were included. The first 188 patients were used to generate ROC curves to define the optimal cut-off values for both Thomas-plots. With the following 348 patients included we studied the performance of the alternative and classical Thomas-plots compared to current anemia diagnostics. RESULTS: Cut-off values were defined (Ret-He: 31.2 pg, sTfR/log(fer): 0.91, and Tf/log(fer): 1.71). With both Thomas-plots the amount of e causa ignota (ECI) cases dropped from 39% to 27%. A more in depth analysis on the iron status of anemia of chronic disease (ACD) patients and a subdivision between latent and classical iron deficiencies could be made with the help of both plots. A shift from classical iron deficiency anemia (IDA) cases according to the classical Thomas-plot toward functional IDA according to the alternative Thomas-plot was observed. CONCLUSION: The alternative Thomas-plot is an effective tool that gives a more in depth view on the iron status of anemic patients. In addition, it is easier to implement due to the use of transferrin rather than the soluble transferrin receptor.