Daun Jeong, Minyoung Choi, Seung Jin Maeng, Hanbeom Yoon, Jong Eun Park, Gun Tak Lee, Sung Yeon Hwang, Tae Gun Shin, Sung Phil Chung, Tae Ho Lim, on behalf of the Korean Shock Society
In Press, Received October 15, 2025 Accepted December 29, 2025 Available online January 14, 2026
Objective Sepsis remains a major clinical challenge because of its complex, heterogeneous, and multidimensional clustering patterns. This study aimed to investigate the association between vasopressor administration and machine learning–derived clusters based on initial vital signs and lactate measurements obtained in emergency department (ED) and intensive care unit (ICU) settings.
Methods A retrospective cohort analysis was performed using data from the Korean Shock Society Septic Shock (KOSS) Registry (septic shock in the ED) and the Marketplace for Medical Information in Intensive Care (MIMIC)-IV database (ICU patients with suspected infection). To derive clusters, k-means clustering was applied to six initial vital signs and serum lactate measurements. The primary outcome was vasopressor administration. Secondary outcomes included second vasopressor administration and 28-day mortality.
Results A total of 17,500 patients were included in the analysis (KOSS cohort, n=7,130; MIMIC-IV cohort, n=10,370). K-means clustering identified three distinct clusters in each cohort. In the KOSS cohort, Cluster 3 was characterized by the lowest mean arterial pressure (MAP) (62 mmHg [IQR, 53–71]) and the highest diastolic shock index (DSI) (2.6 [2.3–3.0]). This cluster was associated with the highest rates of vasopressor administration (93.9%), second vasopressor administration (33.5%), and 28-day mortality (25.3%) (all p<0.001). Comparable physiological and clinical patterns were observed in the MIMIC-IV cohort, in which Cluster 3 likewise demonstrated the lowest MAP (68 mmHg [60–76]) and highest DSI (2.0 [1.8–2.3]). This group similarly exhibited the poorest outcomes, including vasopressor administration (41.0%), second vasopressor administration (16.7%), and 28-day mortality (29.0%).
Conclusion Machine learning–derived clusters based on initial vital signs and serum lactate levels demonstrated different patterns of vasopressor use and mortality. The clinical utility of this approach for guiding timely or targeted vasopressor therapy requires prospective validation.
Sean Hickey, Kusum S. Mathews, Jennifer Siller, Judah Sueker, Mitali Thakore, Deepa Ravikumar, Ruben E Olmedo, Jolion McGreevy, Roopa Kohli-Seth, Brendan Carr, Evan S. Leibner
Clin Exp Emerg Med 2020;7(4):319-325. Published online December 31, 2020
The coronavirus disease 2019 (COVID-19) pandemic mandated rapid, flexible solutions to meet the anticipated surge in both patient acuity and volume. This paper describes one institution’s emergency department (ED) innovation at the center of the COVID-19 crisis, including the creation of a temporary ED–intensive care unit (ICU) and development of interdisciplinary COVID-19–specific care delivery models to care for critically ill patients. Mount Sinai Hospital, an urban quaternary academic medical center, had an existing five-bed resuscitation area insufficiently rescue due to its size and lack of negative pressure rooms. Within 1 week, the ED-based observation unit, which has four negative pressure rooms, was quickly converted into a COVID-19–specific unit, split between a 14-bed stepdown unit and a 13-bed ED-ICU unit. An increase in staffing for physicians, physician assistants, nurses, respiratory therapists, and medical technicians, as well as training in critical care protocols and procedures, was needed to ensure appropriate patient care. The transition of the ED to a COVID-19–specific unit with the inclusion of a temporary expanded ED-ICU at the beginning of the COVID-19 pandemic was a proactive solution to the growing challenges of surging patients, complexity, and extended boarding of critically ill patients in the ED. This pandemic underscores the importance of ED design innovation with flexible spacing, interdisciplinary collaborations on structure and services, and NP ventilation systems which will remain important moving forward.
Citations
Citations to this article as recorded by
Leveraging Observation Units for Disaster Response: Cases of Efficient Care Delivery in Disaster Incidents Iyesatta M. Emeli, Mallika Singh, Esther Hwang, Irfan Husain, Christopher Caspers, Alexander Isakov, Michael A. Ross JACEP Open.2026; 7(2): 100307. CrossRef
State of the Art: Observation Units in the Emergency Department, an Interim Practice Update and Policy Review Christopher Caspers, Mallika Singh, Christopher W. Baugh, Lauren T. Southerland, Jason J. Bischof, Andrew Weinberger, Hardeep Singh, George B. Hughes, Matthew Wheatley, Michael A. Ross JACEP Open.2026; 7(3): 100354. CrossRef
Critical Care Delivery in the Emergency Department Won-Jun Kuk, Jun Soo Park, Kyle J. Gunnerson Critical Care Clinics.2024; 40(3): 497. CrossRef
Risk factors for converting traditional wards to temporary intensive care units during the COVID‐19 pandemic: Insights from nurses' perspectives Wenyu Li, Xiuli Lin, Zhenhong Fang, Xufei Fang, Xiuyun Zheng, Wenyu Tu, Xiaofang Feng Nursing in Critical Care.2024; 29(6): 1412. CrossRef
Strengthening the ICUs' human resource‐related responses to Covid‐19: A rapid review of the experience during the first year of public health emergency Aizhan Tursunbayeva, Stefano Di Lauro The International Journal of Health Planning and Management.2023; 38(1): 22. CrossRef
Early discharge from maternity ward in response to the COVID-19 pandemic: Impact on emergency attendance M. Ducros, P. Tourneux, C. Fontaine Archives de Pédiatrie.2023; 30(1): 25. CrossRef
Prehospital emergency medical service utilization and associated factors among critically ill COVID-19 patients treated at centers in Addis Ababa, Ethiopia Ararso Baru Olani, Lemlem Beza, Menbeu Sultan, Tariku Bekelcho, Michael Alemayehu, Collins Otieno Asweto PLOS Global Public Health.2023; 3(2): e0001158. CrossRef
Impact of COVID-19 outbreak on acute gallbladder disease in the emergency department Dal Sakong, Michael Sung Pil Choe, Woo Young Nho, Chang Won Park Clinical and Experimental Emergency Medicine.2023; 10(1): 84. CrossRef
Critical Care Anywhere: A Novel Emergency Critical Care Consult Service in a Rural Health Network Katelin Morrissette, Skyler Lentz, Ramsey Herrington, Mariah McNamara, Jada Barton, William E. Baker NEJM Catalyst.2023;[Epub] CrossRef
Training and redeployment of healthcare workers to intensive care units (ICUs) during the COVID-19 pandemic: a systematic review Norha Vera San Juan, Sigrún Eyrúnardóttir Clark, Matthew Camilleri, John Paul Jeans, Alexandra Monkhouse, Georgia Chisnall, Cecilia Vindrola-Padros BMJ Open.2022; 12(1): e050038. CrossRef
Perspectives of nursing directors on emergency nurse deployment during the pandemic of COVID‐19: A nationwide cross‐sectional survey in mainland China Ling‐xiao He, Hong‐fei Ren, Feng‐jiao Chen, Zhong‐lan Chen, Cong Wang, Rui‐xue Zhang, Yan Jiang Journal of Nursing Management.2022; 30(5): 1147. CrossRef
Seventeen Years Is Too Long to Move From the ICU to the Emergency Department* Brook Danboise, Khalid Sherani, David J. Wallace Critical Care Medicine.2022; 50(6): 1032. CrossRef
Managing bottlenecks in the perioperative setting: Optimizing patient care and reducing costs Maks Mihalj, Andrea Corona, Lukas Andereggen, Richard D. Urman, Markus M. Luedi, Corina Bello Best Practice & Research Clinical Anaesthesiology.2022; 36(2): 299. CrossRef
Associations between the COVID-19 Pandemic and Hospital Infrastructure Adaptation and Planning—A Scoping Review Costase Ndayishimiye, Christoph Sowada, Patrycja Dyjach, Agnieszka Stasiak, John Middleton, Henrique Lopes, Katarzyna Dubas-Jakóbczyk International Journal of Environmental Research and Public Health.2022; 19(13): 8195. CrossRef
The effect of COVID-19 pandemic on the length of stay and outcomes in the emergency department Soh Yeon Chun, Ho Jung Kim, Han Bit Kim Clinical and Experimental Emergency Medicine.2022; 9(2): 128. CrossRef
Endotracheal Intubation Using C-MAC Video Laryngoscope vs. Direct Laryngoscope While Wearing Personal Protective Equipment Da Saem Kim, Daun Jeong, Jong Eun Park, Gun Tak Lee, Tae Gun Shin, Hansol Chang, Taerim Kim, Se Uk Lee, Hee Yoon, Won Chul Cha, Yong Jin Sim, Song Yi Park, Sung Yeon Hwang Journal of Personalized Medicine.2022; 12(10): 1720. CrossRef
Impact of ICU strain on outcomes Abhijit Duggal, Kusum S. Mathews Current Opinion in Critical Care.2022; 28(6): 667. CrossRef
Impact of COVID-19 Pandemic on Management and Outcomes in Patients with Septic Shock in the Emergency Department Daun Jeong, Gun Tak Lee, Jong Eun Park, Tae Gun Shin, Kyunga Kim, Doeun Jang, Won Young Kim, You Hwan Jo, Sung Phil Chung, Jin Ho Beom, Sung-Hyuk Choi, Woon Yong Kwon, Gil Joon Suh, Byuk Sung Ko, Kap Su Han, Jong Hwan Shin, Hanjin Cho, Sung Yeon Hwang Journal of Personalized Medicine.2022; 12(11): 1803. CrossRef
Predictors of Respiratory Support Use in Emergency Department Patients With COVID-19-Related Respiratory Failure Neha N Goel, Erin Eschbach, Daniel McConnell, Bryan Beattie, Sean Hickey, John Rozehnal, Evan Leibner, Gary Oldenburg, Kusum S Mathews Respiratory Care.2022; 67(9): 1091. CrossRef
Outcomes for in-hospital cardiac arrest for COVID-19 patients at a rural hospital in Southern California Rahul V. Nene, Nicole Amidon, Christian A. Tomaszewski, Gabriel Wardi, Andrew Lafree The American Journal of Emergency Medicine.2021; 47: 244. CrossRef
Objective There is a traditional assumption that to maximize stroke volume, the point beneath which the left ventricle (LV) is at its maximum diameter (P_max.LV) should be compressed. Thus, we aimed to derive and validate rules to estimate P_max.LV using anteroposterior chest radiography (chest_AP), which is performed for critically ill patients urgently needing determination of their personalized P_max.LV.
Methods A retrospective, cross-sectional study was performed with non-cardiac arrest adults who underwent chest_AP within 1 hour of computed tomography (derivation:validation=3:2). On chest_AP, we defined cardiac diameter (CD), distance from right cardiac border to midline (RB), and cardiac height (CH) from the carina to the uppermost point of left hemi-diaphragm. Setting point zero (0, 0) at the midpoint of the xiphisternal joint and designating leftward and upward directions as positive on x- and y-axes, we located P_max.LV (x_max.LV, y_max.LV). The coefficients of the following mathematically inferred rules were sought: x_max.LV=α0*CD-RB; y_max.LV=β0*CH+γ0 (α0: mean of [x_max.LV+RB]/CD; β0, γ0: representative coefficient and constant of linear regression model, respectively).
Results Among 360 cases (52.0±18.3 years, 102 females), we derived: x_max.LV=0.643*CD-RB and y_max.LV=55-0.390*CH. This estimated P_max.LV (19±11 mm) was as close as the averaged P_max.LV (19±11 mm, P=0.13) and closer than the three equidistant points representing the current guidelines (67±13, 56±10, and 77±17 mm; all P<0.001) to the reference identified on computed tomography. Thus, our findings were validated.
Conclusion Personalized P_max.LV can be estimated using chest_AP. Further studies with actual cardiac arrest victims are needed to verify the safety and effectiveness of the rule.
Citations
Citations to this article as recorded by
Development of artificial intelligence-driven biosignal-sensitive cardiopulmonary resuscitation robot Taegyun Kim, Gil Joon Suh, Kyung Su Kim, Hayoung Kim, Heesu Park, Woon Yong Kwon, Jaeheung Park, Jaehoon Sim, Sungmoon Hur, Jung Chan Lee, Dong Ah Shin, Woo Sang Cho, Byung Jun Kim, Soyoon Kwon, Ye Ji Lee Resuscitation.2024; 202: 110354. CrossRef
Optimal Landmark for Chest Compressions during Cardiopulmonary Resuscitation Derived from a Chest Computed Tomography in Arms-Down Position Pimpan Usawasuraiin, Borwon Wittayachamnankul, Boriboon Chenthanakij, Juntima Euathrongchit, Phichayut Phinyo, Theerapon Tangsuwanaruk Journal of Cardiovascular Development and Disease.2022; 9(4): 100. CrossRef
Hand Placement During Chest Compressions in Parturients: A Pilot Study to Identify the Location of the Left Ventricle Using Transthoracic Echocardiography C. Delgado, K. Dawson, B. Schwaegler, R. Zachariah, S. Einav, L. Bollag Obstetric Anesthesia Digest.2021; 41(2): 84. CrossRef
Optimum chest compression point might be located rightwards to the maximum diameter of the right ventricle: A preliminary, retrospective observational study Hyoungouk Kim, Sung‐Bin Chon, Seung Min Yoo, Himchan Choi, Kwang‐Yeol Park Acta Anaesthesiologica Scandinavica.2020; 64(7): 1002. CrossRef
First
Prev
