Penglin Ma1, Jingtao Liu2, Feng Shen3, Xuelian Liao4, Ming Xiu5, Heling Zhao6, Mingyan Zhao7, Jing Xie8, Peng Wang9, Man Huang10, Tong Li11, Meili Duan12, Kejian Qian13, Yue Peng14, Feihu Zhou15, Xin Xin16, Xianyao Wan17, ZongYu Wang18, Shusheng Li19, Jianwei Han20, Zhenliang Li21, Guolei Ding22, Qun Deng23, Jicheng Zhang24, Yue Zhu25, Wenjing Ma26, Jingwen Wang27, Yan Kang4, Zhongheng Zhang28. 1. Department of Critical Care Medicine, Guiqian International General Hospital, Guiyang, People's Republic of China. 2. Department of Critical Care Medicine, The 8th Medical Center of Chinese, PLA General Hospital, Beijing, 100091, People's Republic of China. 3. Department of Intensive Care Unit, Guizhou Medical University Affiliated Hospital, Guiyang, People's Republic of China. 4. Department of Critical Care Medicine, West China Hospital of Sichuan University, Chengdu, People's Republic of China. 5. Department of Intensive Care Unit, The First Hospital of Jilin University, Changchun, People's Republic of China. 6. Department of Critical Care Medicine, Hebei General Hospital, Shijiazhuang, People's Republic of China. 7. Department of Critical Care Medicine, The First Affiliated Hospital of Harbin Medical University, Harbin, People's Republic of China. 8. General Intensive Care Unit Department, The First Affiliated Hospital of Fujian Medical University, Fuzhou, People's Republic of China. 9. Department of Critical Care Medicine, Fu Xing Hospital, Capital Medical University, Beijing, People's Republic of China. 10. General Intensive Care Unit, Second Affiliated Hospital of Zhejiang University, Hangzhou, People's Republic of China. 11. Department of Critical Care Medicine, Beijing Tongren Hospital, Capital Medical University, Beijing, People's Republic of China. 12. Department of Critical Care Medicine, Beijing Friendship Hospital, Capital Medical University, Beijing, People's Republic of China. 13. Department of Critical Care Medicine, The First Affiliated Hospital of Nanchang University, Nanchang, People's Republic of China. 14. Department of Critical Care Medicine, The Third Xiangya Hospital, Central South University, Changsha, People's Republic of China. 15. Department of Critical Care Medicine, Chinese PLA General Hospital, Beijing, People's Republic of China. 16. Surgical Intensive Care Unit, Beijing Chao-Yang Hospital, Capital Medical University, Beijing, People's Republic of China. 17. The First Affiliated Hospital of Dalian Medical University, Dalian, People's Republic of China. 18. Department of Intensive Care, Peking University Third Hospital, Beijing, People's Republic of China. 19. Department of Emergency, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, People's Republic of China. 20. Department of Critical Care Medicine, The 8th medical Center of Chinese, PLA General Hospital, Beijing, People's Republic of China. 21. Department of Critical Care, Beijing PingGu Hospital, Capital Medical University, Beijing, People's Republic of China. 22. Intensive Care Unit, The Hospital of Shunyi District, Beijing, People's Republic of China. 23. Department of Critical Care Medicine, The 4th Medical Center of Chinese, PLA General Hospital, Beijing, People's Republic of China. 24. Department of Critical Care Medicine, Shandong Provincial Hospital, Affiliated to Shandong First Medical University, Jinan, People's Republic of China. 25. Department of Critical Care, Beijing Luhe Hospital, Capital Medical University, Beijing, People's Republic of China. 26. Department of Critical Care, Beijing Miyun Hospital, Beijing, People's Republic of China. 27. Intensive Care Unit, Beijing Changping District Hospital, Beijing, People's Republic of China. 28. Department of Emergency Medicine, Sir Run Run Shaw Hospital, Zhejiang University School of Medicine, Hangzhou, 310016, People's Republic of China. zh_zhang1984@zju.edu.cn.
Abstract
BACKGROUND: Septic shock comprises a heterogeneous population, and individualized resuscitation strategy is of vital importance. The study aimed to identify subclasses of septic shock with non-supervised learning algorithms, so as to tailor resuscitation strategy for each class. METHODS: Patients with septic shock in 25 tertiary care teaching hospitals in China from January 2016 to December 2017 were enrolled in the study. Clinical and laboratory variables were collected on days 0, 1, 2, 3 and 7 after ICU admission. Subclasses of septic shock were identified by both finite mixture modeling and K-means clustering. Individualized fluid volume and norepinephrine dose were estimated using dynamic treatment regime (DTR) model to optimize the final mortality outcome. DTR models were validated in the eICU Collaborative Research Database (eICU-CRD) dataset. RESULTS: A total of 1437 patients with a mortality rate of 29% were included for analysis. The finite mixture modeling and K-means clustering robustly identified five classes of septic shock. Class 1 (baseline class) accounted for the majority of patients over all days; class 2 (critical class) had the highest severity of illness; class 3 (renal dysfunction) was characterized by renal dysfunction; class 4 (respiratory failure class) was characterized by respiratory failure; and class 5 (mild class) was characterized by the lowest mortality rate (21%). The optimal fluid infusion followed the resuscitation/de-resuscitation phases with initial large volume infusion and late restricted volume infusion. While class 1 transitioned to de-resuscitation phase on day 3, class 3 transitioned on day 1. Classes 1 and 3 might benefit from early use of norepinephrine, and class 2 can benefit from delayed use of norepinephrine while waiting for adequate fluid infusion. CONCLUSIONS: Septic shock comprises a heterogeneous population that can be robustly classified into five phenotypes. These classes can be easily identified with routine clinical variables and can help to tailor resuscitation strategy in the context of precise medicine.
BACKGROUND:Septic shock comprises a heterogeneous population, and individualized resuscitation strategy is of vital importance. The study aimed to identify subclasses of septic shock with non-supervised learning algorithms, so as to tailor resuscitation strategy for each class. METHODS:Patients with septic shock in 25 tertiary care teaching hospitals in China from January 2016 to December 2017 were enrolled in the study. Clinical and laboratory variables were collected on days 0, 1, 2, 3 and 7 after ICU admission. Subclasses of septic shock were identified by both finite mixture modeling and K-means clustering. Individualized fluid volume and norepinephrine dose were estimated using dynamic treatment regime (DTR) model to optimize the final mortality outcome. DTR models were validated in the eICU Collaborative Research Database (eICU-CRD) dataset. RESULTS: A total of 1437 patients with a mortality rate of 29% were included for analysis. The finite mixture modeling and K-means clustering robustly identified five classes of septic shock. Class 1 (baseline class) accounted for the majority of patients over all days; class 2 (critical class) had the highest severity of illness; class 3 (renal dysfunction) was characterized by renal dysfunction; class 4 (respiratory failure class) was characterized by respiratory failure; and class 5 (mild class) was characterized by the lowest mortality rate (21%). The optimal fluid infusion followed the resuscitation/de-resuscitation phases with initial large volume infusion and late restricted volume infusion. While class 1 transitioned to de-resuscitation phase on day 3, class 3 transitioned on day 1. Classes 1 and 3 might benefit from early use of norepinephrine, and class 2 can benefit from delayed use of norepinephrine while waiting for adequate fluid infusion. CONCLUSIONS:Septic shock comprises a heterogeneous population that can be robustly classified into five phenotypes. These classes can be easily identified with routine clinical variables and can help to tailor resuscitation strategy in the context of precise medicine.
Authors: Koceila Bouferrache; Jean-Bernard Amiel; Loïc Chimot; Vincent Caille; Cyril Charron; Philippe Vignon; Antoine Vieillard-Baron Journal: Crit Care Med Date: 2012-10 Impact factor: 7.598
Authors: Donald M Yealy; John A Kellum; David T Huang; Amber E Barnato; Lisa A Weissfeld; Francis Pike; Thomas Terndrup; Henry E Wang; Peter C Hou; Frank LoVecchio; Michael R Filbin; Nathan I Shapiro; Derek C Angus Journal: N Engl J Med Date: 2014-03-18 Impact factor: 91.245
Authors: Timothy E Sweeney; Tej D Azad; Michele Donato; Winston A Haynes; Thanneer M Perumal; Ricardo Henao; Jesús F Bermejo-Martin; Raquel Almansa; Eduardo Tamayo; Judith A Howrylak; Augustine Choi; Grant P Parnell; Benjamin Tang; Marshall Nichols; Christopher W Woods; Geoffrey S Ginsburg; Stephen F Kingsmore; Larsson Omberg; Lara M Mangravite; Hector R Wong; Ephraim L Tsalik; Raymond J Langley; Purvesh Khatri Journal: Crit Care Med Date: 2018-06 Impact factor: 7.598
Authors: Sandra L Peake; Anthony Delaney; Michael Bailey; Rinaldo Bellomo; Peter A Cameron; D James Cooper; Alisa M Higgins; Anna Holdgate; Belinda D Howe; Steven A R Webb; Patricia Williams Journal: N Engl J Med Date: 2014-10-01 Impact factor: 91.245
Authors: R Phillip Dellinger; Mitchell M Levy; Jean M Carlet; Julian Bion; Margaret M Parker; Roman Jaeschke; Konrad Reinhart; Derek C Angus; Christian Brun-Buisson; Richard Beale; Thierry Calandra; Jean-Francois Dhainaut; Herwig Gerlach; Maurene Harvey; John J Marini; John Marshall; Marco Ranieri; Graham Ramsay; Jonathan Sevransky; B Taylor Thompson; Sean Townsend; Jeffrey S Vender; Janice L Zimmerman; Jean-Louis Vincent Journal: Crit Care Med Date: 2008-01 Impact factor: 7.598
Authors: Paul R Mouncey; Tiffany M Osborn; G Sarah Power; David A Harrison; M Zia Sadique; Richard D Grieve; Rahi Jahan; Sheila E Harvey; Derek Bell; Julian F Bion; Timothy J Coats; Mervyn Singer; J Duncan Young; Kathryn M Rowan Journal: N Engl J Med Date: 2015-03-17 Impact factor: 91.245
Authors: Zudin A Puthucheary; Jochen S Gensichen; Aylin S Cakiroglu; Richard Cashmore; Lara Edbrooke; Christoph Heintze; Konrad Neumann; Tobias Wollersheim; Linda Denehy; Konrad F R Schmidt Journal: Crit Care Date: 2020-09-25 Impact factor: 9.097
Authors: Zhongheng Zhang; Lin Chen; Ping Xu; Qing Wang; Jianjun Zhang; Kun Chen; Casey M Clements; Leo Anthony Celi; Vitaly Herasevich; Yucai Hong Journal: NPJ Digit Med Date: 2022-07-19
Authors: Longxiang Su; Yansheng Li; Shengjun Liu; Siqi Zhang; Xiang Zhou; Li Weng; Mingliang Su; Bin Du; Weiguo Zhu; Yun Long Journal: Front Med (Lausanne) Date: 2022-04-14