Shock (Augusta, Ga.) · 2025
A New Proposed Approach to Categorize Sepsis Sub-phenotypes Based on Combined Trajectories of Perfusion Index and Central Venous Oxygen Saturation.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Wang Xinchen, Wang Xiaoting, Kang Yan, Zhou Xiang, Long Yun, Du Wei
Аннотация
BACKGROUND: Circulatory disturbances in sepsis are heterogeneous but may manifest as a mismatch between peripheral perfusion and tissue oxygenation. Peripheral perfusion index (PI) and central venous oxygen saturation (ScvO 2 ) have been independently validated as prognostic markers in sepsis; however, their combined utility remains unexplored.
METHODS: This retrospective cohort study included patients admitted to three intensive care units between January 1, 2015, and December 31, 2021. A novel index, PISO = (PI - 1.1) × ScvO 2 , was calculated every 6 hours during the first 72 hours following sepsis onset. Dynamic time warping and hierarchical agglomerative clustering were applied to PISO trajectories to identify sub-phenotypes, with internal validation and sensitivity analysis. Machine learning algorithms were used to predict the sub-phenotypes using clinical features. Survival was assessed using Kaplan-Meier analysis, with particular focus on the association with 12-hour cumulative fluid balance.
RESULTS: Among 12,975 patients, 5,205 (40.1%) developed sepsis; 720 patients with complete PI and ScvO 2 data were included. Three distinct PISO sub-phenotypes were identified. Phenotype 3 (n = 190, 33.0%), characterized by a persistently declining trajectory from negative, was associated with the poorest clinical outcomes. Random forest classifier demonstrated the best performance and identified key early clinical features differentiating each sub-phenotype. Kaplan-Meier analysis revealed that a 12-hour cumulative fluid balance ≤0 mL was significantly associated with improved 28-day survival in phenotype 1 ( P = 0.038) and phenotype 2 ( P = 0.033).
CONCLUSIONS: A persistently negative and declining PISO index within the first 72 hours of sepsis onset identifies a high-risk sub-phenotype. A nonpositive 12-hour cumulative fluid balance is associated with improved survival in select sub-phenotypic groups.
METHODS: This retrospective cohort study included patients admitted to three intensive care units between January 1, 2015, and December 31, 2021. A novel index, PISO = (PI - 1.1) × ScvO 2 , was calculated every 6 hours during the first 72 hours following sepsis onset. Dynamic time warping and hierarchical agglomerative clustering were applied to PISO trajectories to identify sub-phenotypes, with internal validation and sensitivity analysis. Machine learning algorithms were used to predict the sub-phenotypes using clinical features. Survival was assessed using Kaplan-Meier analysis, with particular focus on the association with 12-hour cumulative fluid balance.
RESULTS: Among 12,975 patients, 5,205 (40.1%) developed sepsis; 720 patients with complete PI and ScvO 2 data were included. Three distinct PISO sub-phenotypes were identified. Phenotype 3 (n = 190, 33.0%), characterized by a persistently declining trajectory from negative, was associated with the poorest clinical outcomes. Random forest classifier demonstrated the best performance and identified key early clinical features differentiating each sub-phenotype. Kaplan-Meier analysis revealed that a 12-hour cumulative fluid balance ≤0 mL was significantly associated with improved 28-day survival in phenotype 1 ( P = 0.038) and phenotype 2 ( P = 0.033).
CONCLUSIONS: A persistently negative and declining PISO index within the first 72 hours of sepsis onset identifies a high-risk sub-phenotype. A nonpositive 12-hour cumulative fluid balance is associated with improved survival in select sub-phenotypic groups.