Association of dynamic trajectories of inflammation-coagulation panel biomarkers with acute kidney injury in pregnancy-related sepsis: a group-based multi-trajectory modeling study

Pregnancy-related sepsis is a major cause of preventable maternal mortality, with acute kidney injury (AKI) as a severe complication driven by dysregulated inflammation-coagulation crosstalk. Conventional static single-time-point biomarkers fail to capture dynamic disease progression and patient heterogeneity, limiting precise AKI risk stratification. This study aimed to identify empirically derived trajectory classes based on dynamic inflammation-coagulation biomarker patterns in pregnancy-related sepsis and evaluate their associations with post-landmark AKI. A single-center retrospective cohort study was conducted on 306 ICU-admitted patients with pregnancy-related sepsis. Group-based multi-trajectory modeling (GBMTM) was performed using 72-hour longitudinal data of white blood cell count, platelet count, prothrombin time, and fibrinogen. The primary analysis was restricted to the 253 patients who were alive, still hospitalized, and AKI-free at the 72-hour landmark (the landmark cohort). The primary outcome was post-landmark incident AKI that occurred after 72 h of ICU admission, defined according to the KDIGO AKI criteria. Kaplan-Meier analysis, confounder-adjusted Cox regression, competing risk model, ROC analysis, Boruta feature ranking, subgroup analyses, and seven sensitivity analyses were applied for statistical validation. Five trajectory classes were identified with acceptable classification reliability (overall average posterior probability 0.934, entropy 0.888). In the landmark cohort ( n = 253), post-landmark AKI occurred in 55 patients (21.7%), with a marked risk gradient across classes: 6.0% in the lowest-risk class (Class A) versus 45.0% in the highest-risk class (Class E). After adjustment for ten prespecified confounders, all four lower-risk classes were associated with lower post-landmark AKI hazard compared with Class E (adjusted HRs 0.10–0.39, all P < 0.05). The five-class trajectory label showed only weak discriminatory performance for post-landmark AKI (optimism-corrected AUC 0.614), and exploratory Boruta analysis ranked the trajectory label highest among candidate variables. This study identified five empirically derived inflammation-coagulation trajectory classes in pregnancy-related sepsis that show a marked gradient in post-landmark AKI risk in this single-center cohort. However, the weak discrimination and single-center retrospective design indicate that these findings are exploratory and hypothesis-generating. The generalizability and clinical utility of this trajectory classification require validation in large multicenter prospective cohorts.

Authors

Institutions

Publication Details

Journal
BMC Pregnancy and Childbirth
Published
2026-11-17
DOI
https://doi.org/10.1186/s12884-026-10225-x
Primary Topic
Acute Kidney Injury Research
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Association of dynamic trajectories of inflammation-coagulation panel biomarkers with acute kidney injury in pregnancy-related sepsis: a group-based multi-trajectory modeling study

Xuefei Hou, 李佳蓉, Suru Yue, Jia Wang et al.
BMC Pregnancy and Childbirth
Acute Kidney Injury Research
article

Association of dynamic trajectories of inflammation-coagulation panel biomarkers with acute kidney injury in pregnancy-related sepsis: a group-based multi-trajectory modeling study

Xuefei Hou, 李佳蓉, Suru Yue, Jia Wang, Xiaocong Sun, Jing Liao, Yuyang Wu, Jiayuan Wu
article en

Abstract

Pregnancy-related sepsis is a major cause of preventable maternal mortality, with acute kidney injury (AKI) as a severe complication driven by dysregulated inflammation-coagulation crosstalk. Conventional static single-time-point biomarkers fail to capture dynamic disease progression and patient heterogeneity, limiting precise AKI risk stratification. This study aimed to identify empirically derived trajectory classes based on dynamic inflammation-coagulation biomarker patterns in pregnancy-related sepsis and evaluate their associations with post-landmark AKI. A single-center retrospective cohort study was conducted on 306 ICU-admitted patients with pregnancy-related sepsis. Group-based multi-trajectory modeling (GBMTM) was performed using 72-hour longitudinal data of white blood cell count, platelet count, prothrombin time, and fibrinogen. The primary analysis was restricted to the 253 patients who were alive, still hospitalized, and AKI-free at the 72-hour landmark (the landmark cohort). The primary outcome was post-landmark incident AKI that occurred after 72 h of ICU admission, defined according to the KDIGO AKI criteria. Kaplan-Meier analysis, confounder-adjusted Cox regression, competing risk model, ROC analysis, Boruta feature ranking, subgroup analyses, and seven sensitivity analyses were applied for statistical validation. Five trajectory classes were identified with acceptable classification reliability (overall average posterior probability 0.934, entropy 0.888). In the landmark cohort ( n = 253), post-landmark AKI occurred in 55 patients (21.7%), with a marked risk gradient across classes: 6.0% in the lowest-risk class (Class A) versus 45.0% in the highest-risk class (Class E). After adjustment for ten prespecified confounders, all four lower-risk classes were associated with lower post-landmark AKI hazard compared with Class E (adjusted HRs 0.10–0.39, all P < 0.05). The five-class trajectory label showed only weak discriminatory performance for post-landmark AKI (optimism-corrected AUC 0.614), and exploratory Boruta analysis ranked the trajectory label highest among candidate variables. This study identified five empirically derived inflammation-coagulation trajectory classes in pregnancy-related sepsis that show a marked gradient in post-landmark AKI risk in this single-center cohort. However, the weak discrimination and single-center retrospective design indicate that these findings are exploratory and hypothesis-generating. The generalizability and clinical utility of this trajectory classification require validation in large multicenter prospective cohorts.

BMC Pregnancy and Childbirth
Guangdong Medical College (CN), Affiliated Hospital of Guangdong Medical College Hospital (CN)
Acute Kidney Injury Research
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