300 Section - Lactate as a Predictor of Mortality in Ischemic Stroke: Infection versus Non-Infection Groups from the MIMIC-IV Cohort

Wednesday, May 20, 2026

6:00 PM - 7:30 PM PT

Presenter(s)

Qing Mei Wang, MD, PhD

Director, Stroke Biological Recovery Laboratory
Spaulding Rehabilitation Hospital/Harvard Medical School
Boston, Massachusetts, United States

Presenting Author/Autor expositor(s)

MZ

Mingchao Zhou, MD

Research scholar
Spaulding Rehabilitation Hospital; Shenzhen Second People’s Hospital
Boston, Massachusetts, United States

Co-Author/Coautor(s)

JD

Jiao Deng, MS

Attending Physician
Changzhou Hospital of Traditional Chinese Medicine
Changzhou, Jiangsu, China (People's Republic)
LL

Li Luo, PhD

Professor
Soochow University
Suzhou, Jiangsu, China (People's Republic)
XF

Xuehui Fan, MD

Attending Physician
Shenzhen Second People’s Hospital
Shenzhen, Guangdong, China (People's Republic)
YW

Yulong Wang, MS

Professor
Shenzhen Second People’s Hospital
Shenzhen, Guangdong, China (People's Republic)

Corresponding Author/Autor a quien dirigir la corr(s)

Xun Luo, MD (he/him/his)

Academician, European Academy of Natural Sciences; Academy Master Supervisor, Wenzhou Medical University
Wenzhou Medical University Big Health Development Institute
Woburn, Massachusetts, United States

Objectives : Previous studies have suggested that lactate may serve as a prognostic indicator in patients with ischemic stroke. This study aimed to investigate the differential predictive value of lactate for in-hospital mortality among patients with ischemic stroke stratified by infection status.

Design: Data from patients with ischemic stroke were extracted from the MIMIC-IV database. Clinical characteristics, along with baseline lactate and laboratory parameters measured within the first 24 hours of admission, were collected and analyzed. Least Absolute Shrinkage and Selection Operator regression identified key predictors of in-hospital mortality. Kaplan–Meier survival curve and multivariable Cox models assessed the impact of lactate. Accelerated failure time-based path analysis evaluated its direct and indirect effects on mortality.

Results: A total of 469 patients were included in this study, comprising 215 non-infected patients (11 in-hospital deaths) and 254 infected patients (75 in-hospital deaths). Admission lactate ≥2 mmol/L was associated with lower in-hospital survival versus < 2 mmol/L (log-rank p = 0.021 in non-infected and 0.002 in infected patients). Lactate remained an independent predictor of mortality in both groups (HR = 2.961, p < 0.001; HR = 1.191, p = 0.017, respectively). Path analysis showed that lactate had a significant direct effect (DE) on survival time in both non-infected (DE = –0.238, p = 0.001) and infected patients (DE = –0.174, p = 0.002), with an additional indirect effect (IE) associated with white blood cell in infected patients (IE = –0.124, p = 0.008).

Conclusion: The independent association of elevated lactate on admission with in-hospital mortality in patients with ischemic stroke may be interpreted differently depending on infection status. In non-infected patients, timely metabolic optimization may confer prognostic benefits. Conversely, in infected patients, comprehensive management of the interaction between metabolic dysregulation and infection may help to prevent adverse outcomes.