Preceding organ trajectories guide dynamic reassessment after apparent stabilisation in sepsis

Apparent organ recovery can conceal residual vulnerability in sepsis. Across 19,690 adults in four intensive care databases, we evaluated six-hour deterioration in low-severity organs after transition to single-organ dysfunction. At 2,492 eICU reassessment alerts, adding preceding organ trajectories increased capture from 219 to 282 episodes (28.8%; difference 63, 95% CI 38.0 to 93.5). Reassessment depended on the availability and updating of that history. Withholding half of record batches lost 108.3 captures; adaptation recovered approximately one third, and targeted retrieval increased timely correct nominations from 151 to 163. New measurements modified the value of earlier history, while local calibration and alert scheduling improved external reassessment. Among 813 norepinephrine-treated patients, history improved hypotension-or-death discrimination (AUROC 0.5925 to 0.6929) and, in the fixed-model comparison, concentrated larger estimated recorded exposure reductions across treatment priorities (AUTOC difference 25.68 micrograms/kg, 95% CI 5.23 to 46.16). These findings support retaining organ trajectories for dynamic reassessment and treatment prioritisation.

Authors

Institutions

Publication Details

Journal
medRxiv
Published
2026-11-02
DOI
https://doi.org/10.64898/2026.11.01.26367016
Primary Topic
Sepsis Diagnosis and Treatment
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
preprint

Preceding organ trajectories guide dynamic reassessment after apparent stabilisation in sepsis

Xihui Gao, Jian Zhang, David Luk, Hongmei Zi et al.
medRxiv
Sepsis Diagnosis and Treatment
preprint

Preceding organ trajectories guide dynamic reassessment after apparent stabilisation in sepsis

Xihui Gao, Jian Zhang, David Luk, Hongmei Zi, Lu Zheng, Hengbing Zhou, Yiming Sun, Jialin Song, Shuying Chen, Marina Vereshchako, Qingyu Zhou, Hongyong He, Chuhan Zhong, Rongyang Zhou, Luomeng Wang, Jiajun Zhu, Xinran Law
preprint en

Abstract

Apparent organ recovery can conceal residual vulnerability in sepsis. Across 19,690 adults in four intensive care databases, we evaluated six-hour deterioration in low-severity organs after transition to single-organ dysfunction. At 2,492 eICU reassessment alerts, adding preceding organ trajectories increased capture from 219 to 282 episodes (28.8%; difference 63, 95% CI 38.0 to 93.5). Reassessment depended on the availability and updating of that history. Withholding half of record batches lost 108.3 captures; adaptation recovered approximately one third, and targeted retrieval increased timely correct nominations from 151 to 163. New measurements modified the value of earlier history, while local calibration and alert scheduling improved external reassessment. Among 813 norepinephrine-treated patients, history improved hypotension-or-death discrimination (AUROC 0.5925 to 0.6929) and, in the fixed-model comparison, concentrated larger estimated recorded exposure reductions across treatment priorities (AUTOC difference 25.68 micrograms/kg, 95% CI 5.23 to 46.16). These findings support retaining organ trajectories for dynamic reassessment and treatment prioritisation.

medRxiv
Shanghai Medical College of Fudan University (CN), Fudan University (CN), Zhongshan Hospital (CN), Pudong Medical Center (CN), Huashan Hospital (CN)
Sepsis Diagnosis and Treatment
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon
Preceding organ trajectories guide dynamic reassessment after apparent stabilisation in sepsis — Xihui Gao, Jian Zhang, et al. · medRxiv (2026) | TGRS Research Map | TGRS