Alejandro Rodríguez1,2, Manuel Ruiz-Botella3, Ignacio Martín-Loeches4, María Jimenez Herrera5, Jordi Solé-Violan6, Josep Gómez3, María Bodí7,8, Sandra Trefler7, Elisabeth Papiol9, Emili Díaz10, Borja Suberviola11, Montserrat Vallverdu12, Eric Mayor-Vázquez13, Antonio Albaya Moreno14, Alfonso Canabal Berlanga15, Miguel Sánchez16, María Del Valle Ortíz17, Juan Carlos Ballesteros18, Lorena Martín Iglesias19, Judith Marín-Corral20, Esther López Ramos21, Virginia Hidalgo Valverde22, Loreto Vidaur Vidaur Tello23, Susana Sancho Chinesta24, Francisco Javier Gonzáles de Molina25, Sandra Herrero García26, Carmen Carolina Sena Pérez27, Juan Carlos Pozo Laderas28, Raquel Rodríguez García29, Angel Estella30, Ricard Ferrer9. 1. ICU Hospital Universitario Joan XXIII/IISPV/URV, Mallafre Guasch 4, 43007, Tarragona, Spain. ahr1161@yahoo.es. 2. CIBERESUCICOVID, Barcelona, Spain. ahr1161@yahoo.es. 3. Tarragona Health Data Research Working Group (THeDaR), ICU Hospital Universitario Joan XXIII, Tarragona, Spain. 4. Department of Intensive Care Medicine, Multidisciplinary Intensive Care Research Organization (MICRO), St. James's Hospital, Dublin, Ireland. 5. Dean Nursing Faculty, Universitat Rovira i Virgili, Tarragona, Spain. 6. ICU Hospital Universitario Dr. Negrín, Las Palmas de Gran Canaria, Spain. 7. ICU Hospital Universitario Joan XXIII/IISPV/URV, Mallafre Guasch 4, 43007, Tarragona, Spain. 8. CIBERESUCICOVID, Barcelona, Spain. 9. ICU Hospital Universitario Vall d'Hebron, Barcelona, Spain. 10. ICU Hospital Parc Tauli, Sabadell, Spain. 11. ICU Hospital Marqués de Valdecilla, Santander, Spain. 12. ICU Hospital Universitario Arnau de Vilanova, Lleida, Spain. 13. ICU Hospital Verge de la Cinta, Tortosa, Spain. 14. ICU Hospital Universitario de Guadalajara, Guadalajara, Spain. 15. ICU Hospital de La Princesa, Madrid, Spain. 16. ICU Hospital Clinico San Carlos, Madrid, Spain. 17. ICU Hospital Universitario de Burgos, Burgos, Spain. 18. ICU Hospital Clínico de Salamanca, Salamanca, Spain. 19. ICU Hospital Universitario Central de Asturias, Oviedo, Spain. 20. ICU Hospital del Mar, Barcelona, Spain. 21. ICU Hospital Príncipe de Asturias, Alcalá de Henares, Spain. 22. ICU Hospital Complejo Asistencial de Segovia, Segovia, Spain. 23. ICU Hospital Universitario de Donostia, Donosia, Spain. 24. ICU Hospital Universitario y Politécnico La Fe, Valencia, Spain. 25. ICU Hospital Universitario de Terrasa, Terrasa, Spain. 26. ICU Hospital Clínico Universitario Lozano Blesa, Zaragoza, Spain. 27. ICU Hospital Nuestra Señora del Prado, Talavera de la Reina, Spain. 28. ICU Hospital Universitario Reina Sofía, Córdoba, Spain. 29. ICU Complejo Hospitalario Universitario a Coruña, A Coruña, Spain. 30. ICU Hospital Universitario de Jerez, Jerez de la Frontera, Spain.
Abstract
BACKGROUND: The identification of factors associated with Intensive Care Unit (ICU) mortality and derived clinical phenotypes in COVID-19 patients could help for a more tailored approach to clinical decision-making that improves prognostic outcomes. METHODS: Prospective, multicenter, observational study of critically ill patients with confirmed COVID-19 disease and acute respiratory failure admitted from 63 ICUs in Spain. The objective was to utilize an unsupervised clustering analysis to derive clinical COVID-19 phenotypes and to analyze patient's factors associated with mortality risk. Patient features including demographics and clinical data at ICU admission were analyzed. Generalized linear models were used to determine ICU morality risk factors. The prognostic models were validated and their performance was measured using accuracy test, sensitivity, specificity and ROC curves. RESULTS: The database included a total of 2022 patients (mean age 64 [IQR 5-71] years, 1423 (70.4%) male, median APACHE II score (13 [IQR 10-17]) and SOFA score (5 [IQR 3-7]) points. The ICU mortality rate was 32.6%. Of the 3 derived phenotypes, the A (mild) phenotype (537; 26.7%) included older age (< 65 years), fewer abnormal laboratory values and less development of complications, B (moderate) phenotype (623, 30.8%) had similar characteristics of A phenotype but were more likely to present shock. The C (severe) phenotype was the most common (857; 42.5%) and was characterized by the interplay of older age (> 65 years), high severity of illness and a higher likelihood of development shock. Crude ICU mortality was 20.3%, 25% and 45.4% for A, B and C phenotype respectively. The ICU mortality risk factors and model performance differed between whole population and phenotype classifications. CONCLUSION: The presented machine learning model identified three clinical phenotypes that significantly correlated with host-response patterns and ICU mortality. Different risk factors across the whole population and clinical phenotypes were observed which may limit the application of a "one-size-fits-all" model in practice.
BACKGROUND: The identification of factors associated with Intensive Care Unit (ICU) mortality and derived clinical phenotypes in COVID-19patients could help for a more tailored approach to clinical decision-making that improves prognostic outcomes. METHODS: Prospective, multicenter, observational study of critically illpatients with confirmed COVID-19 disease and acute respiratory failure admitted from 63 ICUs in Spain. The objective was to utilize an unsupervised clustering analysis to derive clinical COVID-19 phenotypes and to analyze patient's factors associated with mortality risk. Patient features including demographics and clinical data at ICU admission were analyzed. Generalized linear models were used to determine ICU morality risk factors. The prognostic models were validated and their performance was measured using accuracy test, sensitivity, specificity and ROC curves. RESULTS: The database included a total of 2022 patients (mean age 64 [IQR 5-71] years, 1423 (70.4%) male, median APACHE II score (13 [IQR 10-17]) and SOFA score (5 [IQR 3-7]) points. The ICU mortality rate was 32.6%. Of the 3 derived phenotypes, the A (mild) phenotype (537; 26.7%) included older age (< 65 years), fewer abnormal laboratory values and less development of complications, B (moderate) phenotype (623, 30.8%) had similar characteristics of A phenotype but were more likely to present shock. The C (severe) phenotype was the most common (857; 42.5%) and was characterized by the interplay of older age (> 65 years), high severity of illness and a higher likelihood of development shock. Crude ICU mortality was 20.3%, 25% and 45.4% for A, B and C phenotype respectively. The ICU mortality risk factors and model performance differed between whole population and phenotype classifications. CONCLUSION: The presented machine learning model identified three clinical phenotypes that significantly correlated with host-response patterns and ICU mortality. Different risk factors across the whole population and clinical phenotypes were observed which may limit the application of a "one-size-fits-all" model in practice.
Entities:
Keywords:
Machine learning; Phenotypes; Prognosis; Risk factors; Severe SARS-CoV-2 infection
Authors: Christopher W Seymour; Jason N Kennedy; Shu Wang; Chung-Chou H Chang; Corrine F Elliott; Zhongying Xu; Scott Berry; Gilles Clermont; Gregory Cooper; Hernando Gomez; David T Huang; John A Kellum; Qi Mi; Steven M Opal; Victor Talisa; Tom van der Poll; Shyam Visweswaran; Yoram Vodovotz; Jeremy C Weiss; Donald M Yealy; Sachin Yende; Derek C Angus Journal: JAMA Date: 2019-05-28 Impact factor: 56.272
Authors: Matt Arentz; Eric Yim; Lindy Klaff; Sharukh Lokhandwala; Francis X Riedo; Maria Chong; Melissa Lee Journal: JAMA Date: 2020-04-28 Impact factor: 56.272
Authors: Giacomo Grasselli; Alberto Zangrillo; Alberto Zanella; Massimo Antonelli; Luca Cabrini; Antonio Castelli; Danilo Cereda; Antonio Coluccello; Giuseppe Foti; Roberto Fumagalli; Giorgio Iotti; Nicola Latronico; Luca Lorini; Stefano Merler; Giuseppe Natalini; Alessandra Piatti; Marco Vito Ranieri; Anna Mara Scandroglio; Enrico Storti; Maurizio Cecconi; Antonio Pesenti Journal: JAMA Date: 2020-04-28 Impact factor: 56.272
Authors: J L Serrano-Martínez; J F Machado-Casas; M Redondo-Orts; F Manzano-Manzano; J Castaño-Pérez; J M Pérez-Villares Journal: Med Intensiva (Engl Ed) Date: 2020-06-17
Authors: Elie Azoulay; Jan de Waele; Ricard Ferrer; Thomas Staudinger; Marta Borkowska; Pedro Povoa; Katerina Iliopoulou; Antonio Artigas; Stefan J Schaller; Manu Shankar-Hari; Mariangela Pellegrini; Michael Darmon; Jozef Kesecioglu; Maurizio Cecconi Journal: Crit Care Date: 2020-08-05 Impact factor: 9.097
Authors: Pratik Sinha; Kevin L Delucchi; Daniel F McAuley; Cecilia M O'Kane; Michael A Matthay; Carolyn S Calfee Journal: Lancet Respir Med Date: 2020-01-13 Impact factor: 30.700
Authors: Marwan Saad; Kevin F Kennedy; Hafiz Imran; David W Louis; Ernie Shippey; Athena Poppas; Kenneth E Wood; J Dawn Abbott; Herbert D Aronow Journal: JAMA Date: 2021-11-16 Impact factor: 56.272
Authors: Benjamin L Ranard; Murad Megjhani; Kalijah Terilli; Kevin Doyle; Jan Claassen; Michael R Pinsky; Gilles Clermont; Yoram Vodovotz; Shadnaz Asgari; Soojin Park Journal: Front Med (Lausanne) Date: 2021-11-11
Authors: P Vidal-Cortés; E Díaz Santos; E Aguilar Alonso; R Amezaga Menéndez; M Á Ballesteros; M A Bodí; M L Bordejé Laguna; J Garnacho Montero; M García Sánchez; M López Sánchez; I Martín-Loeches; A Ochagavía Calvo; P Ramírez Galleymore; S Alcántara Carmona; D Andaluz Ojeda; O Badallo Arébalo; H Barrasa González; M Borges Sa; Á Castellanos-Ortega; Á Estella; R Ferrer Roca; V Fraile Gutiérrez; M Fuset Cabanes; C Giménez-Esparza Vich; C González Iglesias; A Hernández-Tejedor; J C Igeño Cano; D Iglesias Posadilla; J J Jiménez Rivera; C Llanos Jorge; J A Llompart-Pou; V López Camps; C Lorencio Cárdenas; P Marcos Neira; M C Martín Delgado; M Martín-Macho González; L Martín Villén; X Nuvials Casals; A Ortiz Suñer; M Quintana Díaz; P Rascado Sedes; M Recuerda Núñez; L Del Río Carbajo; M Rodríguez Aguirregabiria; A Rodríguez Oviedo; I Seijas Betolaza; C Soriano Cuesta; B Suberviola Cañas; C Vera Ching; Á Vidal González; L Zapata Fenor; R Zaragoza Crespo Journal: Med Intensiva Date: 2021-09-16 Impact factor: 2.491
Authors: Sivasubramanium V Bhavani; Philip A Verhoef; Cheryl L Maier; Chad Robichaux; William F Parker; Andre Holder; Rishikesan Kamaleswaran; May D Wang; Matthew M Churpek; Craig M Coopersmith Journal: Crit Care Med Date: 2022-02-01 Impact factor: 9.296