Discussion
The main finding of our study is that among patients with COVID-19, 3 clinical phenotypes were derived using habitual clinical and laboratory variables at ICU admission. The ability of identifying phenotypes using a small set of variables is a crucial step towards clinical application and has important implications for possible differential treatment guided by phenotypes and validated prognostic scoring systems [
18,
19].
Our C phenotype was associated with more than double the ICU mortality than each of the remaining two phenotypes. This C phenotype was characterized by the interplay of older age (> 65 year), a high severity (APACHE II > 15 and SOFA > 5), greater burden of risk factors (hematologic disease and coronary disease) and a higher likelihood of developing further complications (shock and AKI).
Previous studies have implemented clustering techniques to analyze various data sources relating to demographic, geographic, environment, and socioeconomic determinants of health and disease. There are studies that have evaluated treatment decisions and characterized clinical phenotypes associated with complications, ICU admission and mortality risk in critically ill COVID-19 patients. According to the Situation Report & Public Health Guidance published by Johns Hopkins University on March 19th, 2020, people over 60 and those with chronic health conditions are at the highest risk for COVID-19 complications [
20]. To our knowledge, this is the first study with a high number of critically ill patients to analyze the presence of phenotypes in patients with SARS-CoV-2 infection. This multicenter cohort study of 2,022 critically ill patients found that 660 patients (32.6%) died at ICU discharge. Our ICU mortality rates was significantly lower as reported in Yang et al. [
4] in Wuhan, China (61.5%), by Myers et al. [
7] in California, USA (50.0%), by Arentz et al. [
6] in Washington, USA (67%) and by Richardson et al. [
5] in New York, USA (78%), but slightly higher to reported by Grasselli et al. [
8] in Lombardy region, Italy (26%). These observed differences in ICU mortality could respond to different healthcare models and important practice variations in the management of severe COVID-19 patients [
11], but it can also depend on the frequency of presentation of the different phenotypes.
In our study, a great variability in model performance and risk factors were observed during cross-validation to choose the best model to use. In addition, we use 2 different techniques for the selection of important variables, one of them is the “classic” approach dependent on the p-value, while the other, a “modern” statistical approaches, is more in line with the new recommendations [
21]. Although the performance of the models was similar, the variables included in each of them are different. This could be related to the presence of a very heterogeneous patient population, which is revealed during random partitioning (80%/20%) validation of each model or by implementing 2 variable selection techniques. In this context, three clinical phenotypes of COVID-19 patients were derived using routinely available clinical data at ICU admission by an unsupervised cluster analysis. The phenotypes were multidimensional, differed in their demographics, laboratory abnormalities, patterns of organ dysfunction, and associated with ICU mortality. In addition, our phenotypes are not similar with groupings or phenotypes of patients performed so far considering only the presence of clinical complications [
22,
23], or the type of ARDS [
24]. Our COVID-19 phenotypes can be identified at the time of the ICU admission, and thus could be useful in facilitating early tailored therapy and improve prognosis.
Only routinely available clinical and laboratory data were used in the clustering models, and the phenotypes were derived from a large observational multicenter cohort to ensure generalizability. Importantly, we have observed that the variables associated with the ICU mortality varied between the global model and the models developed for each phenotype. The discrimination power (AUC) of A and B phenotypes models improved in comparison to the global model. However, for the C phenotype (severe COVID-19 disease), the performance of the model was not superior respect of the global model. The C phenotype was most strongly correlated with abnormal values of biomarkers as well as clinical features of cardiovascular dysfunction, AKI and subsequently a higher ICU mortality. Although the AUC for C phenotype is lower, the relationship between sensitivity and specificity in C phenotype model might be more appropriate. Specificity can sometimes be more important than sensitivity, because confirming that a person does not have the event under study (survival) is more important than detecting if a person has it.
Recently, several authors have proposed different clinical phenotypes of COVID-19 patients [
22‐
24]. Rello et al. [
22] speculated that COVID-19 has five phenotypic presentations based on physiological and clinical features from published studies. Garcia-Vidal et al. [
23] describe the main clinical complications of hospitalized patients with COVID-19 through classification into three pattern groups (inflammatory, co-infection and thrombotic). However, as the authors acknowledge, the cut-off points of the different biomarkers for defining phenotypes have been arbitrary and not scientifically supported. Finally, Gattinoni et al. [
24] proposing two phenotypes for COVID-19 patients, (1) “Type L” characterized by high compliance and low lung recruitablity and (2) “Type H” with low compliance and high lung recruitability as a two “extremes” of a spectrum of respiratory failure in COVID-19 pneumonia. Despite the importance related to clinical experience in each of these approaches, none of these studies have been developed through a machine learning process to determine phenotypes nor have they been tested for validation.
Hypoxemia has been proposed as a marker of severity for the differentiation of phenotypes [
22,
24]. In our study, the PaO
2/FiO
2 relationship at ICU admission was an independent risk factor for ICU mortality in overall multivariate analysis (as a continuous or dichotomized variable), but was only closely associated with ICU mortality in phenotype C. Other variables such as advanced age, serum D-dimer values and the development of AKI were variables more strongly related to ICU mortality in all subgroups or phenotypes analyzed than PaO
2/FiO
2 at ICU admission.
Our results should be interpreted in the context of the study limitations. First, although phenotypes were found to be generalizable in our population, risk factors and characteristics of clinical phenotypes were derived initially from data at ICU admission of multicenter observational study in Spain. However, these risk factors are similar to those that have been reported by other investigators [
4‐
8]. The cross-validation carried out and the high discrimination observed for each of the models built for phenotypes, suggests their applicability to other populations, but it should be examined considering the high variability observed in patients with COVID-19 and in the support measures applied. Second, because missing data were common for some variables included in the clustering models, multiple imputation was used in the primary analysis. However, variables with high missing values were excluded and the missing threshold used was reported elsewhere [
18]. Third, only routinely available clinical data at ICU admission were used to identify risk factors and clinical phenotypes, and the inclusion of other data related to clinical evolution of patients in the ICU could change risk factors or phenotype assignments. However, our objective was to study early risk factors and phenotypes at ICU admission that may allow for early treatment implementation and as a result improve patient outcome. Fourth, although IL-6 is an excellent severity biomarker, we have not been able to include this biomarker in the models because more than 50% of patients had no IL-6 determination upon ICU admission. Although the inclusion of IL-6 in models could modify or improve their performance, we do not consider it appropriate to impute a large number of missing data. In addition, if IL-6 is a biomarker not usually available its inclusion in the models would not have practical application. Finally, we did not collect data on ethnicity or socioeconomic factors. These factors may play a role in the prevalence of pre-existing comorbidities and mortality due to COVID-19. Our findings should be interpreted within the context of the study population and its generalizability to other populations warrant further investigation.
Acknowledgements
To Alexis Garduno for English manuscript edition and to the COVID-19 SEMICYUC Investigators:
Andalucía:
UCI Hospital Universitario Virgen de Valme (Sevilla): Ana Loza; UCI Hospital Quirón (Huelva): Diego Matallana Zapata; UCI Hospital Universitario Puerto Real (Cádiz): Isabel Díaz Torres, Sonia Ibañez Cuadros, María Recuerda Nuñez, Maria Luz Carmona Pérez, Jorge Gómez Ramos, Alba Villares Casas; UCI Hospital Universitario Virgen de la Macarena (Sevilla): María Luisa Cantón, José Javier González Contreras, Helena Pérez Chomón, Nerissa Alvarez Chicote, Alberto Sousa González; UCI Hospital Universitario Reina Sofía (Córdoba): María De Alba Aparicio, Juan Carlos Pozo Laderas; UCI Hospital Universitario de Jerez (Jerez de la Frontera): Angel Estella, Sara moreno Cano. UCI Hospital Infanta Elena (Huelva): Diego Matallana Zapata.
Aragón:
UCI Hospital Nuestra Señora de Gracia (Zaragoza): Ruth Jorge García; UCI Hospital Clínico Universitario Lozano Blesa (Zaragoza): Laura Sánchez Montori, Sandra Herrero García, Paula Abanses Moreno, Carlos Mayordomo García. UCI Hospital General San Jorge(Huesca): Tomás Mallor Bonet, Paula Omedas Bonafonte, Enric Franquesa Gonzalez, Nestor Bueno Vidales, Paula Ocabo Buil, Carlos Serón Arbeloa; UCI Hospital Universitario Miguel Servet(Zaragoza): Isabel Sancho, Pablo Guerrero Ibañez, Pablo Gutierrez, UCI Hospital Obispo Polanco (Teruel): María Concepción Valdovinos, Raquel Canto.UCI Hospital Nuestra Señora de Gracia (Zaragoza): Ruth Jorge García; UCI Hospital Clínico Universitario Lozano Blesa (Zaragoza): Laura Sánchez Montori, Sandra Herrero García, Paula Abanses Moreno, Carlos Mayordomo García. UCI Hospital General San Jorge(Huesca): Tomás Mallor Bonet, Paula Omedas Bonafonte, Enric Franquesa Gonzalez, Nestor Bueno Vidales, Paula Ocabo Buil, Carlos Serón Arbeloa; UCI Hospital Universitario Miguel Servet(Zaragoza): Isabel Sancho, Pablo Guerrero Ibañez, Pablo Gutierrez, UCI Hospital Obispo Polanco (Teruel): María Concepción Valdovinos, Raquel Canto.
Asturias:
UCI Hospital Universitario San Agustín (Avilés): Ana Luz Balán Mariño, María José Gutiérrez Fernández, Marta Martín Cuadrado, Belén García Arias; UCI Hospital Universitario Central de Asturias (Oviedo): Lorena Forcelledo Espina, Lucía Viña Soria, Lorena Martín Iglesias, Lucía López Amor, Elisabet Fernández Rey, Emilio García Prieto. UCI Hospital Cabueñes (Gijón): Débora Fernández Ruíz, Carla Martínez González.
Baleares:
UCI hospital Universitario Son Llatzer (Palma de Mallorca): Lorenzo Socias, Marcio Borges-Sá, María Aranda Pérez, Antonia Socias. UCI Hospital Quirón Salud Palmaplanas (Palma de Mallorca): José Mª Bonell Goytisolo, Inmaculada Alcalde Mayayo, Carlos Corradini, Isabel Ceniceros, Edwin Rodríguez; UCI Hospital Universitario Son Espases (Palma de Mallorca): Jose Ignacio Ayestarán Rota, Mariana Andrea Novo Novo, Joaquim Colomina Climent, Albert Figueras Castilla, Tomàs Leal Rullan, Maria Magdalena Garcias Sastre; UCI Hospital Comarcal d’Inca(Inca): Rossana Pérez Senoff; UCI Hospital Mateu Orfila (Mao): Ramón Fernández-Cid Bouza.
Canarias:
UCI Complejo Hospitalario Universitario Insular—Materno Infantil (Las Palmas de G.C): Juan Carlos Martín González, Carmen Pérez Ortiz, José Luciano Cabrera Santana, Juan José Cáceres Agra, Domingo González Romero, Ana Casamitjana Ortega; UCI Hospital General de la Palma (Tenerife): Luis Alberto Ramos Gómez, Carolina Montelongo Ojeda; UCI Hospital Universitario Dr. Negrín (Las Palmas de G.C): Jordi Solé-Violán.
Cataluña:
UCI Hospital Universitari de Tarragona Joan XXIII (Tarragona): Alejandro Rodríguez, María Bodí, Gerard Moreno, Sandra Trefler, Laura Claverias, Raquel Carbonell, Erika Esteve, Montserrat Olona, Xavier Teixidó. UCI Hospital Universitari Arnau de Vilanova (Lleida): Monserrat Vallverdú Vidal, Begoña Balsera Garrido. UCI Hospital Universitari Vall d’Hebron (Barcelona): Elisabeth Papiol Gallofré, Raquel Albertos Martell, Rosa Alcaráz Peñarrocha, Xavier Nuvials Casals, Ricard Ferrer Roca; UCI Hospital Verge de la Cinta (Tortosa): Eric Adrián Mayor Vázquez, Ferrán Roche Campo, Pablo Concha Martínez, Diego Franch Llasat;UCI Hospital del Mar (Barcelona): Joan Ramón Masclanz, Judith Marín-Corral, Purificación Pérez, Rosana Muñoz, Clara Vila; UCI Hospital Mutua de Terrasa (Terrasa): Francisco Javier González de Molina, Elisabeth Navas Moya, Josep Trenado; UCI Hospital Sant Joan (Reus): Imma Vallverdú, Eric Castañé; UCI Hospital Parc Tauli (Sabadell): Emili Díaz Santos, Gemma Goma, Edgar Moglia.
Cantabria
UCI Hospital Universitario Marqués de Valdecillas(Santander): Borja Suberviola.
Castilla La Mancha:
UCI Hospital Universitario de Guadalajara (Guadalajara): Antonio Albaya Moreno, Carlos Marian Crespo. UCI Hospital Nuestra Señora del Prado (Toledo): Carmen Carolina Sena Pérez, Francisca Arbol Linde.UCI Hospital Universitario de Guadalajara (Guadalajara): Antonio Albaya Moreno, Carlos Marian Crespo. UCI Hospital Nuestra Señora del Prado (Toledo): Carmen Carolina Sena Pérez, Francisca Arbol Linde.
Castilla y León:
UCI Hospital Virgen de la Concha (Zamora): Diana Monge Donaire, Vega Losada Martínez, Nuria Rodrigo Castroviejo, Gerardo Ferrigno, Reyes Beltrán, Carolina Sanmartino, Concepción Tarancón Maján, Alfredo Marcos Gutiérrez; UCI Complejo Asistencial de Segovia(Segovia): Virginia Hidalgo Valverde, Caridad Martín López; UCI Hospital universitario de Burgos (Burgos): Oihane Badallo, María del Valle Ortiz, Rebeca Vara Arlanzón, David Iglesias Posadilla; UCI Hospital Clínica de Salamanca (Salamanca): María Teresa Recio, Juan Carlos Ballesteros; UCI Complejo Asistencial Universitario de Palencia (Palencia).
Ceuta
UCI Hospital Universitario de Ceuta: Enrique Laza Laza.
Extremadura
UCI Hospital San Pedro de Alcántara (Cáceres): Elena Gallego Curto, Mª Carmen Sánchez García; UCI Hospital de Mérida(Mérida): Miguel Díaz-Tavora, Rosa Mancha.
Galicia:
UCI Hospital Montecelo (Pontevedra): Ana Ortega Montes, Isabel Gallego Barbachano, Eva Sanmartín Mantiñán. UCI CHUAC A Coruña (A Coruña): María Lourdes Cordero, Raquel María Rodríguez García, Jorge Gámez Zapata, María Gestal Vázquez. UCI Centro Hospital Universitario de Ferrol (Ferrol): María José Castro Orjales, María Isabel Álvarez Diéguez. UCI Hospitalario Clínico Universitario de Santiago (Santiago de Compostela): Carmen Rivero Velasco, Beatriz Lence Massa; REA CHUAC A Coruña (A Coruña): María Gestal Vázquez.UCI Hospital Lucus Augusti (Lugo): Ignacio Martínez Varela.
Huelva:
UCI Hospital Infanta Elena (Huelva): Diego Matallana Zapata.
Madrid:
UCI IFEMA (Madrid): Alberto Hernández Tejedor; UCI Hospital Príncipe de Asturias (Madrid): Esther Mª López Ramos, Laura Alcázar Sánchez Elvira, Rocío Molina Montero, Mª Consuelo Pintado Delgado, María Trascasa Muñoz de la Peña, Yaiza Betania Ortiz de Zárate Ansotegui, Alejandra Acha Aranda, Juan Higuera Lucas; UCI Hospital de la Princesa (Madrid): Juan Antonio Sanchez Giralt, Marta Chicot Llano, Nuria Arevalillo Fernández, Marta Sánchez Galindo, Ricardo Andino Ruiz, Alfonso Canabal Berlanga; UCI Hospital Clinico San Carlos (Madrid): Miguel Sánchez, Mercedes Nieto; UCI Hospital HLA la Moncloa(Madrid): Eduardo Arias Sarmiento, Adoración Bueno Blázquez, Rosa María de la Casa, Fátima Martín, Samuel González López.
Murcia
UCI Hospital Morales Meseguer (Murcia): Elena Martínez Quintana, Bernardo Gil Rueda, Áurea Higon Cañigral, Laura López Gómez, Pablo Safwat Bayoumi Delis, Augusto Montenegro Muore, Ángel Andrés Agamez Luengas; UCI Hospital Clínico Universitario Virgen de la Arrixaca (Murcia): Enriqueta Andreu Soler, Ana Beatriz Pérez Pérez, José Higinio de Gea García, Rubén Jara Rubio, Silvia Sánchez Cámara, Alba Moreno Flores, José Moya Sánchez, Daniel Francisco Pérez Martínez,-Mª Desamparados del Rey Carrión; UCI Hospital Reina Sofía (Murcia): María José Rico Lledó, Juana María Serrano Navarro, Juan Francisco Martín Ruíz, Julián Triviño Hidalgo, África López Ferrer, Isabel Cremades Navalón; UCI Hospital Santa Lucía (Cartagena): Josefa Murcia Payá, JM Allegre Gallego; UCI Hospital Rafael Méndez (Lorca): María del Carmen Lorente; UCI Hospital Universitario Mar Menor (San Javier):Marta Gonsalvez.
Navarra
UCI Hospital Reina Sofía (Tudela): Ruth González Natera, Raquel Garrido López de Murillo, Tania Ojuel Gros,Raquel Flecha Viguera, Isabel López González; UCI Hospital García Orcoyen(Estella-Lizarra): Adriana García Herrera.
País Vasco
UCI Hospital Universitario de Donostia (Donostia): Loreto Vidaur Tello, Maialen Aseguinolaza, Itziar Eguibar.
Sevilla:
UCI Hospital Universitario Virgen de la Macarena(Sevilla): María Luisa Cantón Bulnes, Jose Javier González Contreras, Helena Pérez Chomón, Nerissa Álvarez Chicote, Alberto Sousa González.
Valencia:
UCI Hospital Universitario de La Ribera (Alzira): Asunción Marqués Parra, Sergio García Marti, Alberto Lorenzo Aguilar, Laura Bellver Bosch, Victor Gascón Sanchez, Sonia De la Guía Ortega. UCI Hospital Dr. Peset (Valencia): Martín Parejo Montell, Alberto Belenguer Muncharaz, Hector Hernández Garces, Victor Ramírez Montero, Mónica Crespo Gómez, Verónica Martí Algarra; UCI Hospital Universitari i Politècnic La Fe (Valencia): Susana Sancho Chinesta, Joaquin Arguedas Cervera, Faustino Álvarez Cebrian, Begoña Balerdi Pérez, Rosa Jannone Fores, Javier Botella de Maglia; UCI Hospital Clínico Universitario de Valencia (Valencia): Nieves Carbonell Monleón, Jose Ferreres Franco, Ainhoa Serrano Lazaro, Mar Juan Díaz, María Luisa Blasco Cortés; UCI Hospital Virgen de los Lirios de Alcoy (Alicante): Laura Fayos, Julia Giménez, Gaspar Soriano, Ricardo Navarro. UCI Hospital Arnau de Vilanova (Valencia): Sonia Mas, Elena Bisbal, Laura Albert, Johncard Romero, Juan Fernández Cabreara; UCI Hospital Comarcal de Vinarós (Vinarós): Andrea Ortíz.
Principado de Andorra
ICU Hospital Nostra Señyora de Meritxell (Les Esclades): Antonio Margarit Ribas.