Simple Parameters from Complete Blood Count Predict In-Hospital Mortality in COVID-19
Mattia Bellan, Danila Azzolina, Eyal Hayden, Gianluca Gaïdano, Mario Pirisi, Antonio Acquaviva, Gianluca Aimaretti, Paolo Aluffi Valletti, Roberto Angilletta, Roberto Arioli, Gian Carlo Avanzi, Gianluca Avino, Piero Emilio Balbo, Giulia Baldon, Francesca Baorda, Emanuela Barbero, Alessio Baricich, Michela Barini, Francesco Barone‐Adesi, Sofia Battistini, Michela Beltrame, Matteo Bertoli, Stephanie Bertolin, Marinella Bertolotti, Marta Betti, Flavio Bobbio, Paolo Boffano, Lucio Boglione, Silvio Borrè, Matteo Brucoli, Elisa Calzaducca, Edoardo Cammarata, Vincenzo Cantaluppi, Roberto Cantello, Andrea Capponi, Alessandro Carriero, Giuseppe Francesco Casciaro, Luigi Mario Castello, Federico Ceruti, Guido Chichino, Emilio Chirico, Carlo Cisari, Micol Giulia Cittone, Crizia Colombo, Cristoforo Comi, Eleonora Croce, Tommaso Daffara, Pietro Danna, Françesco Della Corte, Simona De Vecchi, Umberto Dianzani, Davide Di Benedetto, Elia Esposto, Fabrizio Faggiano, Zeno Falaschi, Daniela Ferrante, Alice Ferrero, Ileana Gagliardi, Alessandra Galbiati, Silvia Gallo, Pietro Luigi Garavelli, Clara Ada Gardino, Massimiliano Garzaro, Maria Luisa Gastaldello, Francesco Gavelli, Alessandra Gennari, Greta Maria Giacomini, Irene Giacone, Valentina Giai Via, Francesca Giolitti, Laura Cristina Gironi, Carla Gramaglia, Leonardo Grisafi, Ilaria Inserra, Marco Invernizzi, Marco Krengli, Emanuela Labella, Irene Landi, Raffaella Landi, Ilaria Leone, Veronica Lio, Luca Lorenzini, Antonio Maconi, Mario Malerba, Giulia Francesca Manfredi, María Martelli, Letizia Marzari, Paolo Marzullo, Marco Mennuni, Claudia Montabone, Umberto Morosini, Marco Mussa, Ilaria Nerici, Alessandro Nuzzo, Carlo Olivieri, Samuel Alberto Padelli, Massimiliano Panella, Andrea Parisini, Alessio Paschè, Filippo Patrucco
Abstract
Introduction. The clinical course of Coronavirus Disease 2019 (COVID-19) is highly heterogenous, ranging from asymptomatic to fatal forms. The identification of clinical and laboratory predictors of poor prognosis may assist clinicians in monitoring strategies and therapeutic decisions. Materials and Methods. In this study, we retrospectively assessed the prognostic value of a simple tool, the complete blood count, on a cohort of 664 patients ( <a:math xmlns:a="http://www.w3.org/1998/Math/MathML" id="M1"> <a:mi>F</a:mi> </a:math> 260; 39%, median age 70 (56-81) years) hospitalized for COVID-19 in Northern Italy. We collected demographic data along with complete blood cell count; moreover, the outcome of the hospital in-stay was recorded. Results. At data cut-off, 221/664 patients (33.3%) had died and 453/664 (66.7%) had been discharged. Red cell distribution width (RDW) ( <c:math xmlns:c="http://www.w3.org/1998/Math/MathML" id="M2"> <c:msup> <c:mrow> <c:mi>χ</c:mi> </c:mrow> <c:mrow> <c:mn>2</c:mn> </c:mrow> </c:msup> </c:math> 10.4; <e:math xmlns:e="http://www.w3.org/1998/Math/MathML" id="M3"> <e:mi>p</e:mi> <e:mo><</e:mo> <e:mn>0.001</e:mn> </e:math> ), neutrophil-to-lymphocyte (NL) ratio ( <g:math xmlns:g="http://www.w3.org/1998/Math/MathML" id="M4"> <g:msup> <g:mrow> <g:mi>χ</g:mi> </g:mrow> <g:mrow> <g:mn>2</g:mn> </g:mrow> </g:msup> </g:math> 7.6; <i:math xmlns:i="http://www.w3.org/1998/Math/MathML" id="M5"> <i:mi>p</i:mi> <i:mo>=</i:mo> <i:mn>0.006</i:mn> </i:math> ), and platelet count ( <k:math xmlns:k="http://www.w3.org/1998/Math/MathML" id="M6"> <k:msup> <k:mrow> <k:mi>χ</k:mi> </k:mrow> <k:mrow> <k:mn>2</k:mn> </k:mrow> </k:msup> </k:math> 5.39; <m:math xmlns:m="http://www.w3.org/1998/Math/MathML" id="M7"> <m:mi>p</m:mi> <m:mo>=</m:mo> <m:mn>0.02</m:mn> </m:math> ), along with age ( <o:math xmlns:o="http://www.w3.org/1998/Math/MathML" id="M8"> <o:msup> <o:mrow> <o:mi>χ</o:mi> </o:mrow> <o:mrow> <o:mn>2</o:mn> </o:mrow> </o:msup> </o:math> 87.6; <q:math xmlns:q="http://www.w3.org/1998/Math/MathML" id="M9"> <q:mi>p</q:mi> <q:mo><</q:mo> <q:mn>0.001</q:mn> </q:math> ) and gender ( <s:math xmlns:s="http://www.w3.org/1998/Math/MathML" id="M10"> <s:msup> <s:mrow> <s:mi>χ</s:mi> </s:mrow> <s:mrow> <s:mn>2</s:mn> </s:mrow> </s:msup> </s:math> 17.3; <u:math xmlns:u="http://www.w3.org/1998/Math/MathML" id="M11"> <u:mi>p</u:mi> <u:mo><</u:mo> <u:mn>0.001</u:mn> </u:math> ), accurately predicted in-hospital mortality. Hemoglobin levels were not associated with mortality. We also identified the best cut-off for mortality prediction: a <w:math xmlns:w="http://www.w3.org/1998/Math/MathML" id="M12"> <w:mtext>NL</w:mtext> <w:mtext> </w:mtext> <w:mtext>ratio</w:mtext> <w:mo>></w:mo> <w:mn>4.68</w:mn> </w:math> was characterized by an odds ratio for in-hospital <y:math xmlns:y="http://www.w3.org/1998/Math/MathML" id="M13"> <y:mtext>mortality</y:mtext> <y:mtext> </y:mtext> <y:mfenced open="(" close=")"> <y:mrow> <y:mtext>OR</y:mtext> </y:mrow> </y:mfenced> <y:mo>=</y:mo> <y:mn>3.40</y:mn> </y:math> (2.40-4.82), while the OR for a <cb:math xmlns:cb="http://www.w3.org/1998/Math/MathML" id="M14"> <cb:mtext>RDW</cb:mtext> <cb:mo>></cb:mo> <cb:mn>13.7</cb:mn> </cb:math> % was 4.09 (2.87-5.83); a <eb:math xmlns:eb="http://www.w3.org/1998/Math/MathML" id="M15"> <eb:mtext>platelet</eb:mtext> <eb:mtext> </eb:mtext> <eb:mtext>count</eb:mtext> <eb:mo>></eb:mo> <eb:mn>166,000</eb:mn> </eb:math> /μL was, conversely, protective (OR: 0.45 (0.32-0.63)). Conclusion. Our findings arise the opportunity of stratifying COVID-19 severity according to simple lab parameters, which may drive clinical decisions about monitoring and treatment.