Sepsis remains a leading cause of death in critically ill patients. Host mRNA biomarkers may provide complementary biological information for prognostic assessment in sepsis. This study aimed to integrate host mRNA biomarkers and clinical parameters to develop and compare predictive models for short-term mortality using both conventional and machine learning (ML) approaches. In this prospective ICU study (Dec 2022–Mar 2024), 249 patients meeting Sepsis-3 criteria were initially enrolled, and 198 patients with available host mRNA measurements were included in the final analysis. 35 candidate mRNAs were quantified using droplet digital PCR (ddPCR). After initial transcriptomic screening and LASSO-based feature selection, one immune transcript (PAX5) and four clinical variables (hematologic disease, sex, SOFA score, positive blood culture) were retained for model development. Five models were evaluated for predicting 7-, 14-, and 28-day mortality, including Cox proportional hazards regression, random survival forest (RSF), Deep Survival Analysis (DeepSurv), support vector machine (SVM), and extreme gradient boosting (XGBoost). Model discrimination was assessed using AUCs and compared using DeLong’s test. Internal validation was performed using the bootstrap 0.632 method. Among the 198 patients included in the final analysis, mortality was 18.7

Post-transplant infection caused by multidrug-resistant organisms (MDROs) is a major challenge in heart transplantation, especially in recipients requiring veno-arterial extracorporeal membrane oxygenation (VA-ECMO) before surgery. We describe a 52-year-old man with non-ST-elevation myocardial infarction and refractory cardiogenic shock who required VA-ECMO, intra-aortic balloon pump support, continuous renal replacement therapy, and mechanical ventilation before urgent heart transplantation. Before transplantation, he had active pneumonia. Donor respiratory culture grew Acinetobacter baumannii, Pseudomonas aeruginosa, and Staphylococcus aureus, whereas donor blood culture and blood metagenomic next-generation sequencing (mNGS) were negative. After transplantation, serial mNGS and conventional cultures revealed rapidly progressive polymicrobial infection involving Stenotrophomonas maltophilia, Burkholderia multivorans, carbapenem-resistant Acinetobacter baumannii, carbapenem-resistant Klebsiella pneumoniae, and vancomycin-resistant Enterococcus faecium. Antimicrobial therapy was repeatedly adjusted, and VA-ECMO was successfully discontinued on postoperative day 13. However, on postoperative day 16, the patient developed bowel ischemia with gastrointestinal perforation, followed by feculent peritonitis, persistent septic shock, progressive multiorgan dysfunction syndrome, and death on postoperative day 24. This case shows that perioperative infection control in VA-ECMO-bridged urgent heart transplant recipients requires more than broad-spectrum antimicrobial escalation. It requires careful assessment of preoperative infection controllability, interpretation of mNGS in conjunction with culture-based susceptibility testing, and early investigation of occult abdominal ischemia when clinical deterioration is unexplained.

Background The platelet-to-lymphocyte ratio (PLR), a readily available marker that integrates systemic inflammatory burden and immune competence, has emerged as a cost-effective prognostic biomarker in critical care medicine. Elevated PLR has been linked to adverse outcomes across a spectrum of critical illnesses, yet its utility in predicting prognosis among patients with septic acute kidney injury (AKI) remains undefined. Moreover, little is known about how in-hospital trajectories of PLR influence survival outcomes. Method This retrospective study employed data from the Medical Information Mart for Intensive Care IV and West China Hospital of Sichuan University. The primary endpoints were 28-day and 90-day all-cause mortality. The association between baseline PLR/changes in PLR (ΔPLR) and 28-day and 90-day mortality was investigated by survival analysis and Cox proportional hazards models. ΔPLR was calculated as PLR at discharge minus PLR at admission. Patients were stratified based on optimal PLR cut-off values determined from the training cohort, and results were validated internally and externally. Predefined subgroup analyses probed for effect modification across key clinical populations, and mediation analysis was conducted to quantify intermediate variables linking PLR/ΔPLR dynamics to patient outcomes. Result A total of 1,478 patients were included in the baseline-PLR cohort and 982 in the ΔPLR cohort. In the training set, an elevated baseline PLR (≥335.44) was associated with significantly higher 28-day mortality (26.56% vs. 18.66%, P = 0.009) and 90-day mortality (46.48% vs. 35.61%, P = 0.003). These findings were confirmed in both the internal validation cohort and an external cohort, and remained robust after adjustment for demographic, clinical and laboratory confounders. Conversely, lower ΔPLR also predicted the 28-day (14.81% vs. 6.94%, P = 0.007; HR = 2.261, P = 0.005) and 90-day mortality (24.07% vs. 15.41%, P = 0.029; HR = 1.702, P = 0.017), as well as prolonged hospital stay (28.64 vs. 21.64 days, P = 0.004). The associations between PLR or ΔPLR and mortality remained robust after adjusting for confounding factors. Subgroup analyses indicated that the prognostic value of PLR was particularly pronounced in non-urinary tract infection patients, and a low baseline PLR conferred a significant survival benefit in male patients. Mediation analysis revealed that changes in white blood cell count (ΔWBC) mediated 27.99% of the association between ΔPLR and 28-day mortality. Conclusion Both baseline PLR and its dynamic change during hospitalization may serve as significant predictors of mortality in septic AKI. PLR-based indices could aid in risk stratification and early identification of high-risk patients.

Subway systems are vital for urban transit, yet their air contains iron-oxide nanoparticles that may threaten commuters' respiratory and cardiovascular health. Despite their prevalence, the microscopic properties, formation mechanisms and health effect of these particles remain poorly understood. Here we analyze subway-derived dust sampled at four sites along three metro lines in Hangzhou, China, revealing that magnetic nanoparticles constitute a substantial fraction of subway aerosols. These particles exist in two forms: alpha-Fe2O3 from surface dust and airborne Fe3O4 nanoparticles with a magnetic core encapsulated in an amorphous SiO2 shell, primarily originating from wheel-rail and brake-wheel friction. Notably, we detect nanoscale Fe3O4 particles in lung tissues of subway commuters, demonstrating their inhalation and pulmonary deposition in humans. Mouse inhalation exposure experiments further confirm that Fe3O4 nanoparticles can induce pronounced lung injury. Our findings highlight friction-derived magnetic nanoparticles as a potential public health risk and underscore the need for strategies to mitigate commuter exposure in urban subway systems.

BACKGROUND:Mortality from acute myocardial infarction (MI) has declined significantly in the past decade for nondiabetic patients. However, both morbidity and mobility of ischemic heart failure (IHF) persistently escalate in the diabetic population via incompletely understood mechanisms. Recent studies demonstrated that small extracellular vesicles (sEVs) released from nondiabetic and diabetic adipocytes (ADps) exert opposite effects on acute myocardial ischemia and reperfusion (MI/R) injury. However, whether and how ADp sEVs may protect against post-MI remodeling and IHF, and more important, whether and how diabetes may impair this protective effect, remain unknown. METHODS:sEVs were isolated from epididymal fat pads of nondiabetic animals and intramyocardially injected in nondiabetic or diabetic hearts subjected to MI (90 minutes of MI per 4 weeks of reperfusion). RESULTS:sEV treatment significantly attenuated post-MI cardiac remodeling and improved cardiac function in nondiabetic mice. However, the protection was not observed in diabetic hearts. In adult cardiomyocytes isolated from nondiabetic hearts, sEVs rapidly (15 minutes) activated cell salvage kinases (ERK [extracellular signal-regulated kinase], AMPK [AMP-activated protein kinase], and ACC [acetyl-CoA carboxylase]) and suppressed oxidative stress-induced cell death, suggesting sEV external surface molecules are responsible for the observed cytoprotection. The Exo-Flow (a technology detecting sEV external surface molecules) demonstrated that adiponectin (APN) is enriched on the sEV external surface. The sEVs from APN knockout mice or APN neutralization (NU) antibody pretreated sEVs failed to protect the heart against IHF. Moreover, the cardioprotective effects of sEVs were abolished in APN receptor-1 (AdipoR1)-deficient mice (the primary receptor for APN signaling in the heart) or in mice overexpressing GRK2 (G-protein-coupled receptor kinase 2, a kinase that phosphorylates and inactivates AdipoR1). Finally, diabetes significantly increased cardiac GRK2 expression and AdipoR1 phosphorylation, which prevented sEVs from exerting their beneficial effects. Restoring AdipoR1 function by knockin a mutated phosphorylation-resistant AdipoR1 (AdipoR1S205A) via AAV9 (adeno-associated virus 9)-mediated gene delivery rescued ADp sEV cardioprotection in diabetic mice. CONCLUSIONS:Our study reveals that APN is enriched on the ADp-derived external surface of sEVs and is biologically active, playing a critical role in ADp-cardiomyocyte communication. Diabetes disrupts this communication by enhancing GRK2-mediated AdipoR1 phosphorylation, impairing sEV signaling, and exacerbating IHF. These findings provide new insights into the pathophysiology and therapy of IHF in diabetes.

Studies suggest that the colonization of multidrug-resistant organism in the gut of healthcare workers is similar to that of healthy individuals. However, due to exposure to medical environments, is the abundance of antibiotic resistance genes (ARG) in the gut of ICU healthcare workers higher than that of healthy individuals? Prospective, multicentre, cross-sectional study. Eight medical centers in China, recruiting from January 2024 to February 2024. 303 Healthy people (201 ICU healthcare workers and 103 healthy controls) were screened and 290 Healthy people (191 ICU healthcare workers and 99 healthy controls) were included in analysis. Fecal samples were collected and subjected to metagenomic sequencing. We compared the total ARG abundance, ARG diversity, and gut microbiome composition between the two groups. After adjusting for age, sex, and body mass index, ICU healthcare workers exhibited a significantly higher total ARG abundance compared to healthy controls (fold change = 1.22, 95

BACKGROUND:Carbapenem-resistant organisms (CROs) are among the most concerning antimicrobial resistance pathogens, especially in intensive care units (ICUs). The role of environmental contamination in the transmission of CROs has been increasingly recognized. The aim of this study was to establish a replicable strategy on CROs' infection prevention and control (IPC) in ICUs. METHODS:This study was conducted in 4 phases. Environmental surveillance of CROs was conducted in ICUs every month to identify specific risks, and targeted IPC measures were implemented accordingly. Incidence rates of CRO-related healthcare-associated infections (HAIs) were calculated in each phase to assess the effectiveness of the IPC strategies. RESULTS:The detection of CROs in the ICU environment showed a downtrend during the study, with a significant decrease from 3.27% in baseline phase to 1.07% in phase III (P < .05). The detection of CROs in the open-bay ICU environment was significantly higher than that in the single-patient room ICU at baseline (3.97% vs 1.51%, P < .05) and became comparable in phase III after the implementation of our IPC strategies. The incidence of CRO-related HAIs per 1,000 ICU patient-days also decreased significantly from 2.333 in the baseline phase to 1.143 in phase III (P < .001). CONCLUSIONS:Our environmental surveillance-driven CRO IPC strategies were associated with reductions in environmental CRO contamination and CRO-related HAIs in ICUs; however, the results were limited by unmeasured confounders and the quasi-experimental design.

Background: Several predictive models for invasive pulmonary aspergillosis (IPA) based on clinical characteristics have been reported. Nevertheless, the significance of other concurrently detected microorganisms in IPA patients is equally noteworthy. This study aimed to develop a risk prediction model for IPA by integrating clinical and microbiological characteristics. Methods: This retrospective study was conducted in adult intensive care units (ICUs) of 17 medical centers in China. Clinical data were collected from patients with severe pneumonia who underwent clinical metagenomics of bronchoalveolar lavage fluid between January 1, 2019, and June 30, 2023. Subsequently, patients were randomly assigned to training and validation cohorts in a 7:3 ratio. In the training cohort, potential influencing factors were identified through univariate analysis, clinical practice, and existing literature, and a risk prediction model was constructed using multivariate logistic regression analysis. The performance of this model was then assessed and validated in the validation cohort. Results: Out of 1737 patients initially included in the study, 898 were ultimately analyzed, of which 100 (11%) were diagnosed with IPA. The risk prediction model for IPA, incorporating microbiological characteristics, identified six independent risk factors, namely age, immunosuppression, chronic kidney disease, connective tissue disease, liver failure, and cytomegalovirus positivity. The model demonstrated a superior discriminative ability, with area under the curve (AUC) values of 0.791 and 0.792 in the training and validation cohorts, respectively. Sensitivity and specificity reached 73.1% and 74.9%, respectively, and the model demonstrated good calibration. Conclusion: This study developed a novel risk prediction model for IPA incorporating microbiological characteristics based on clinical metagenomics. The model exhibited good discriminative ability and calibration.

BackgroundNon-anticoagulation is a commonly used strategy in continuous renal replacement therapy (CRRT) among patients with high-bleeding risk. However, the optimal blood flow rate (BFR) to maximize filter and circuit life remains uncertain. This study is designed to elucidate the impact of different BFRs on the durability of filters and circuits in CRRT without anticoagulation.MethodsThis single-center, prospective, three-arm, single-blind, randomized controlled trial (RCT) will involve adult patients requiring non-anticoagulation continuous veno-venous hemodiafiltration (CVVHDF). A total of 486 filters and circuits will be enrolled and randomly assigned to one of three BFR groups: low (150 mL/min), medium (200 mL/min), or high (250 mL/min) BFR group. The outcomes will be analyzed by both intention-to-treat analysis and per-protocol analysis. The primary outcome is filter and circuit life, which is defined as the time from CRRT initiation to CRRT termination due to extracorporeal circuit clotting or other reasons, alongside the proportion of patent circuits at 24, 48, and 72 hours. Secondary outcomes encompass clinical outcomes and potential adverse events such as bleeding and hemodynamic alterations.DiscussionThis study is aiming at comparing the filter and circuit life under different BFR levels during CVVHDF without anti-coagulation. The results may add knowledge to the optimal BFR to prevent extracorporeal circulation clotting and prolong filter and circuit life in non-anticoagulation CRRT.Trial registrationThe study has been registered at https://www.chictr.org.cn (ChiCTR2400087819).

BACKGROUND:Aspergillus, the causative pathogen of invasive pulmonary aspergillosis (IPA)-a highly lethal infectious disease-produces spores with diameters that fall within the fine particulate matter (PM2.5) range and are frequently detected in ambient PM2.5 samples. In this study, we investigated whether pre-admission exposure to PM2.5 is associated with an increased risk of IPA in patients with severe pneumonia. METHODS:Daily PM2.5 levels for the six months before admission were obtained from a multicenter retrospective cohort and a multicenter prospective cohort. Multivariable logistic regression was used to assess whether pre-admission PM2.5 exposure is an independent risk factor of IPA in both cohorts. Pooled concentration-response curves determined the dose-response relationship. Mediation analysis was used to assess whether the presence of Aspergillus acts as a mediator between PM2.5 exposure and 28-day mortality. FINDINGS:Among 2287 patients, higher average daily PM2.5 exposure over six-month before admission was independently associated with an increased risk of IPA. For every 10 μg/m3 increase in average daily PM2.5 exposure during the six months before admission, the risk of IPA increased by 21% (95% CI: 10%-32%). The dose-response relationship was linear, and results remained robust across subgroups and sensitivity analyses. Mediation analysis showed that Aspergillus positivity was found to mediate 21.26% (95% CI: 4.5%-48%; P = 0.008) of the relationship between daily PM2.5 exposure in the 90 days preceding admission and 28-day mortality. INTERPRETATION:In this study, we evaluate the association between pre-admission PM2.5 exposure and IPA. Our findings demonstrate that higher concentrations of PM2.5 prior to admission are associated with an increased risk of IPA among ICU-admitted patients with severe pneumonia. In addition, our findings suggest that Aspergillus mediate the association between PM2.5 exposure and mortality in this population. FUNDING:This work was supported by National Science and Technology Major Project of China (2025ZD0549000), National Natural Science Foundation of China (grant no. 82202356, 82341109, and 82173645), the Zhejiang Provincial Natural Science Fund (grant no. LTGY24H190001), "Pioneer" and "Leading Goose" R&D Program of Zhejiang (grant no. 2025C02090).

Sepsis is a critical condition associated with high morbidity and mortality, emphasizing the need for reliable biomarkers for its diagnosis and prognosis. This study uses advanced immunological techniques to evaluate monocytic CD39 (mCD39) expression as a potential marker in sepsis. This prospective observational cohort study included 206 participants from the First Affiliated Hospital, Zhejiang University School of Medicine between April 2022 and September 2023. Participants were categorized into four groups: healthy donors, patients with mild infections, post-cardiac surgery patients (non-infectious inflammation), and sepsis patients. Peripheral Blood Mononuclear Cells were analyzed using mass cytometry time-of-flight (CyTOF) with a 42-marker immune panel and flow cytometry targeting monocytes. Statistical analyses included ROC curves for diagnostic and prognostic performance and Kaplan–Meier survival analysis for prognostic evaluation. Sepsis patients exhibited significantly lower monocytic CD39 expression than mild infection and post-surgery groups (p < 0.05). The diagnostic performance analysis revealed that mCD39 effectively distinguished sepsis from mild infection (AUC = 0.877) and non-infectious inflammation (AUC = 0.935). Prognostic analysis identified low mCD39 expression as a strong predictor of short-term survival, with a 7-day survival AUC of 0.85 (p = 0.037). Kaplan–Meier analysis showed that sepsis patients with low mCD39 expression had significantly lower 28-day survival rates (56.7

Methylprednisolone is still used to treat adults with septic shock in real-world clinical settings, despite current international guidelines recommending hydrocortisone. The aim of this study was to assess the effect of methylprednisolone vs hydrocortisone on 30-day mortality among critically ill patients with septic shock. We conducted a retrospective cohort study on adults with septic shock using the MIMIC-IV v3.0 database. Patients who received methylprednisolone after diagnosis were matched using propensity score matching (PSM) with those received hydrocortisone, to balance confounding factors between groups. The primary outcome was the 30-day mortality rate. Subgroup and sensitivity analyses were performed to assess the robustness of the conclusions. A total of 1,607 septic shock patents were enrolled in this study, with an overall 30-day mortality rate of 42.1

BACKGROUND:Central nervous system (CNS) infections impose a substantial global burden of morbidity and mortality, necessitating accurate and timely diagnosis for optimal clinical management. Metagenomic next-generation sequencing (mNGS) has been demonstrated as a valuable tool for pathogen detection in patients with suspected CNS infections. METHODS:From December 2019 to June 2023, we conducted a comprehensive analysis of 520 cerebrospinal fluid samples collected from patients with suspected intracranial infections across six intensive care units. All pathogen-positive results identified through metagenomic next-generation sequencing were subsequently validated by an independent laboratory. RESULTS:Metagenomic NGS detected 105 microbial species in 520 clinical samples, comprising 64 bacterial species (61.0%), 16 DNA viruses (15.2%), 13 fungal species (12.4%), and 7 RNA viruses (6.7%). The 30-day mortality rate among all study participants was 18.5%. Metagenomic NGS identified 172 infection cases, whereas conventional culture methods detected only 31 cases. For CNS infections, mNGS demonstrated diagnostic sensitivity, specificity, and accuracy rates of 59%, 90.5%, and 72.5%, respectively. CONCLUSIONS:Metagenomic NGS testing facilitates accelerated diagnosis of CNS infections and informs evidence-based clinical management decisions.

OBJECTIVES:The clinical relevance of Epstein-Barr virus (EBV) detection in cerebrospinal fluid (CSF) using metagenomic next-generation sequencing (mNGS) in patients with suspected intracranial infections has not been fully clarified. This study aimed to assess the clinical features, imaging characteristics, and outcomes in EBV-positive patients. METHODS:We retrospectively enrolled patients with suspected intracranial infection who underwent commercial mNGS of CSF samples between December 2019 and October 2023 across 13 hospitals in four cities. Clinical data were collected, and multivariable logistic regression was performed to assess the association between EBV positivity and unfavorable outcomes. RESULTS:Among 507 patients, 51 (10.1%) were EBV-positive in CSF. EBV-positive cases showed higher CSF mononuclear cell proportions (83.8% vs. 32.8%, P < 0.001), lower chloride levels (123.0 vs. 126.1 mmol/l, P = 0.019), and more frequent herpesvirus co-detection (21.6% vs. 8.6%, P = 0.007). Imaging abnormalities did not differ significantly. After adjustment, multivariable logistic regression showed that EBV was not an independent risk factor for clinical outcomes (OR 1.005, 95% CI 0.516-1.972, P = 0.988). CONCLUSIONS:EBV was detected in 10.1% of patients in CSF samples from patients with suspected intracranial infections. EBV positivity correlated with a lymphocytic CSF profile and co-detection of other herpesviruses but was not independently associated with unfavorable outcomes.

Diabetic cardiomyopathy (DCM) is a leading cause of death in diabetic patients. Current therapies do not adequately resolve this problem and focus only on the optimal level of blood glucose for patients. Ferroptosis plays an important role in diabetes mellitus and cardiovascular diseases. However, the role of ferroptosis in DCM remains unclear. Differentially expressed ferroptosis-related genes (DE-FRGs) were identified by intersection of the GSE26887 dataset and the Ferroptosis Database. The associations between the DE-FRGs and immune cells in DCM, estimated via the CIBERSORTx algorithm, were analysed. Flow cytometry (FCM) was used to evaluate the infiltration of immune cells in myocardial tissues. The expression of DE-FRGs, glutathione peroxidase 4 and solute carrier family 7 member 11 was examined via real-time quantitative PCR and Western blotting. Three DE-FRGs were identified: heat shock protein family B (small) member 1 (HSPB1), microsomal glutathione S-transferase 1 (MGST1) and solute carrier family 40 member 1 (SLC40A1), which are closely linked to immune cells in DCM. In vivo, the levels of CD8 + T cells, B cells and regulatory T (Treg) cells were significantly decreased in the DCM group, whereas the levels of CD4 + T cells, M1 cells, M2 cells and monocytes were increased. Diabetes significantly decreased HSPB1 and MGST1 levels and increased ferroptosis compared with the Normal group. Furthermore, the ferroptosis inhibitor ferrostatin-1 (Fer-1) alleviated high-fat diet (HFD)-induced cardiomyocyte injury and rescued ferroptosis. These findings suggest that the ferroptosis-related genes HSPB1 and MGST1 are closely related to immune cell infiltration and may be therapeutic targets for DCM.

Acute kidney injury (AKI) is a clinical syndrome characterized by an abrupt loss of kidney function and is associated with increased morbidity and mortality. Remote ischemic preconditioning (rIPC) is a nonpharmacological intervention involving brief episodes of ischemia in distal tissues, which may provide protection from kidney injury, but its underlying mechanism remain elusive. In a previous study, we demonstrated that NOX4 can serve as a potential therapeutic target in AKI and is associated with the upregulation of inflammation and apoptosis. Therefore, we hypothesized that rIPC might attenuate AKI by inhibiting the NOX4-mediated NF-κB signaling pathway and apoptosis. In this study, we demonstrated that rIPC protected kidney function and pathological injury in lipopolysaccharide (LPS)-induced, cisplatin-induced and ischemic-reperfusion injury (IRI)-induced AKI mouse models. rIPC significantly inhibited the activation of NF-κB and tubular epithelial apoptosis in AKI mice, and hypoxic preconditioning (HPC) similarly suppressed NF-κB and apoptosis of TCMK-1 cells. Notably, rIPC intervention alone slightly increased/preconditioned NOX4 expression in control group mice, while substantially inhibiting NOX4 overexpression when the mice were subjected to AKI insults. Mechanistically, In LPS-stimulated TCMK-1 cells overexpressing NOX4, when treated with rIPC, the excessive activation of NF-κB and apoptosis was further alleviated. These findings demonstrated that rIPC is a potential therapeutic method against AKI and that NOX4 plays a central role in mediating the protective effects of rIPC through the inhibition of NF-κB signaling and tubular apoptosis.

Herpesviruses are widely distributed in the lower respiratory tract, yet no study has comprehensively characterized their clinical features and prognostic impact in severe pneumonia. In this multicenter, retrospective study, we included severe pneumonia patients who underwent bronchoalveolar lavage fluid (BALF) metagenomic testing in intensive care units across 17 medical centers from January 2019 to June 2023. Based on metagenomic results, patients were categorized into herpesvirus-negative, HSV-1, EBV, CMV, HHV-6B, and HHV-7 groups. Propensity score matching and multivariable Cox regression were used to compare mortality between herpesvirus-positive and -negative patients. Interaction analyses were conducted to assess the impact of co-detection of different herpesviruses. Besides, main findings were validated using data from a prospective multicenter cohort. Among 1,737 enrolled patients, the 28-day mortality rate was 41.3

Purpose:Pneumonia is common in ICU patients with neurological dysfunction, but differences in pulmonary pathogen distribution in this population remain unclear. This study aimed to compare pathogen profiles, clinical features, and outcomes between ICU patients with and without neurological dysfunction. Methods:This regional multicenter retrospective study included adult patients with severe pneumonia admitted to intensive care units (ICUs) in 11 hospitals across Zhejiang and Henan Provinces in mainland China between December 2018 and November 2023. All patients required invasive mechanical ventilation and underwent bronchoalveolar lavage fluid metagenomic next-generation sequencing (mNGS). Patients were classified into neurological dysfunction (ND) and without neurological dysfunction (WND) groups. Clinical characteristics, microbiological findings, and outcomes were compared. Propensity score matching (PSM) and Cox regression were used to assess prognosis. Results:Among 1737 patients, 636 (41.8%) were in the ND group. After PSM, the ND group showed a higher 28-day ICU mortality rate and shorter time to death compared to the WND group. However, ND was not identified as an independent risk factor for 28-day mortality in Cox analysis. The prevalence of Acinetobacter baumannii, Klebsiella pneumoniae, Burkholderia, Serratia marcescens, Elizabethkingia, Clostridium spp. and Ureaplasma was higher in ND patients. Significant differences in the prevalence of Haemophilus influenzae, Fusarium oxysporum, Fusobacterium nucleatum, Porphyromonas gingivalis, Mycobacterium abscessus, Escherichia coli, varicella-zoster virus (VZV), Epstein-Barr virus (EBV), and cytomegalovirus (CMV) were also observed. Conclusion:ICU patients with neurological dysfunction exhibited distinct pulmonary pathogen profiles and worse outcomes. These findings may inform empirical antimicrobial strategies. Further prospective studies are warranted to validate these results.