Sepsis remains a leading cause of critical illness worldwide, yet the underlying mechanisms driving its profound and persistent immune dysfunction are incompletely understood. The bone marrow, as the birthplace of all immune cells, plays a central role in orchestrating systemic immune responses. Emerging evidence from animal models suggests that sepsis triggers emergency myeloid-biased hematopoiesis in the bone marrow, characterized by expansion of myeloid progenitors and myeloid-derived suppressor cells (MDSCs) at the expense of lymphoid and erythroid lineages. This bone marrow remodeling precedes peripheral immune alterations and may represent the initiating event of sepsis-induced immunosuppression. However, direct clinical evidence in humans is scarce. This prospective, single-center cohort study aims to systematically characterize bone marrow hematopoietic remodeling in patients with septic shock, compared to critically ill non-septic patients and healthy volunteers, and to determine whether the degree of myeloid lineage bias correlates with disease severity, immunosuppression, and adverse clinical outcomes. This study will enroll three cohorts. Bone marrow aspirates and peripheral blood samples will be collected at 48-72 hours post-enrollment for flow cytometric immunophenotyping of hematopoietic stem/progenitor cells, MDSC subsets, and PD-L1 expression, as well as cytokine profiling and exploratory single-cell transcriptomics. Rectal swabs will be collected synchronously for 16S rRNA sequencing and untargeted metabolomics to investigate the association between gut microbiota, microbial metabolites, and bone marrow myeloid skewing, testing the gut-bone marrow-immune axis hypothesis. Clinical severity (SOFA/APACHE II), secondary infections, and 90-day mortality will be assessed to evaluate prognostic value. By integrating bone marrow hematopoiesis, gut microbiome, and clinical outcomes, this study seeks to provide novel mechanistic insights into sepsis-induced immunoparalysis and identify potential biomarkers or therapeutic targets for immune restoration.
Study Type
OBSERVATIONAL
Enrollment
45
Bone marrow aspiration was performed at the posterior superior iliac spine under local anesthesia 24-48 hours after enrollment. Using a standard aspirate needle and strict aseptic technique, approximately 2-3 mL of bone marrow aspirate was collected.
Percentage and Absolute Count of Hematopoietic Stem Cells (HSCs) in Bone Marrow
HSCs are defined as Lin- CD34⁺ CD38- CD90⁺ CD45RA- cells. The outcome is measured as both (a) the percentage among total bone marrow nucleated cells, and (b) the absolute count per 10⁶ total bone marrow nucleated cells. Comparison is made between the sepsis-associated critical illness cohort and the two control groups (critically ill non-septic cohort and healthy cohort).
Time frame: Between 48 and 72 hours after enrollment (preferably day 3)
Percentage and Absolute Count of Common Myeloid Progenitors (CMPs) in Bone Marrow
CMPs are defined as Lin- CD34⁺ CD38⁺ CD123⁺ CD45RA- cells. The outcome is measured as both (a) the percentage among total bone marrow nucleated cells, and (b) the absolute count. Comparison is made between the sepsis cohort and the two control groups.
Time frame: Between 48 and 72 hours after enrollment (preferably day 3)
Percentage and Absolute Count of Granulocyte-Monocyte Progenitors (GMPs) in Bone Marrow
GMPs are defined as Lin- CD34⁺ CD38⁺ CD123⁺ CD45RA⁺ cells. The outcome is measured as both (a) the percentage among total bone marrow nucleated cells, and (b) the absolute count. This is a core indicator of myeloid lineage expansion. Comparison is made between the sepsis cohort and the two control groups.
Time frame: Between 48 and 72 hours after enrollment (preferably day 3)
Percentage and Absolute Count of Megakaryocyte-Erythroid Progenitors (MEPs) in Bone Marrow
MEPs are defined as Lin- CD34⁺ CD38⁺ CD123- CD45RA- cells. The outcome is measured as both (a) the percentage among total bone marrow nucleated cells, and (b) the absolute count. This indicator reflects erythroid/megakaryocytic lineage suppression. Comparison is made between the sepsis cohort and the two control groups.
Time frame: Between 48 and 72 hours after enrollment (preferably day 3)
Percentage and Absolute Count of Common Lymphoid Progenitors (CLPs) in Bone Marrow
CLPs are defined as Lin- CD34⁺ CD38⁺ CD127⁺ cells. The outcome is measured as both (a) the percentage among total bone marrow nucleated cells, and (b) the absolute count. This indicator reflects lymphoid lineage suppression. Comparison is made between the sepsis cohort and the two control groups.
Time frame: Between 48 and 72 hours after enrollment (preferably day 3)
GMP-to-CLP Ratio and Absolute Differential Count Index in Bone Marrow (Primary Composite Index)
The primary composite index of myeloid lineage bias consists of two parallel metrics: (a) the GMP-to-CLP ratio, calculated as (percentage of GMPs)/(percentage of CLPs); and (b) the absolute differential index, calculated as (absolute count of GMPs)/(absolute count of CLPs). Both metrics quantify myeloid-versus-lymphoid lineage skewing. A higher value indicates greater myeloid bias. Comparison is made between the sepsis cohort and the two control groups.
Time frame: Between 48 and 72 hours after enrollment (preferably day 3)
Percentage and Absolute Count of PMN-MDSCs in Bone Marrow
Polymorphonuclear myeloid-derived suppressor cells (PMN-MDSCs) are defined as CD45⁺ CD11b⁺ CD14- CD15⁺ CD33⁺ HLA-DR-/low cells. The outcome is measured as both (a) the percentage among total bone marrow nucleated cells, and (b) the absolute count. Comparison is made between the sepsis cohort and the two control groups.
Time frame: Between 48 and 72 hours after enrollment (preferably day 3)
Percentage and Absolute Count of M-MDSCs in Bone Marrow
Monocytic myeloid-derived suppressor cells (M-MDSCs) are defined as CD45⁺ CD11b⁺ CD14⁺ HLA-DR-/low CD15- cells. The outcome is measured as both (a) the percentage among total bone marrow nucleated cells, and (b) the absolute count. Comparison is made between the sepsis cohort and the two control groups.
Time frame: Between 48 and 72 hours after enrollment (preferably day 3)
PD-L1 Expression Level and PD-L1⁺ Absolute Count on Bone Marrow MDSCs
PD-L1 (programmed death-ligand 1) expression on both PMN-MDSCs and M-MDSCs is measured as three parallel metrics: (a) the percentage of PD-L1⁺ cells within each MDSC subset; (b) the median fluorescence intensity (MFI) of PD-L1 on these cells; and (c) the absolute count of PD-L1⁺ PMN-MDSCs and PD-L1⁺ M-MDSCs. These indicators reflect the immunosuppressive functional burden of MDSCs. Comparison is made between the sepsis cohort and the two control groups.
Time frame: Between 48 and 72 hours after enrollment (preferably day 3)
Percentage and Absolute Count of T Cells in Bone Marrow
T cells are defined as CD45⁺ CD56- CD3⁺ cells. The outcome is measured as both (a) the percentage among total bone marrow nucleated cells, and (b) the absolute count. Comparison is made between the sepsis cohort and the two control groups.
Time frame: Between 48 and 72 hours after enrollment (preferably day 3)
Percentage and Absolute Count of NK Cells in Bone Marrow
Natural killer (NK) cells are defined as CD45⁺ CD3- CD56⁺ cells. The outcome is measured as both (a) the percentage among total bone marrow nucleated cells, and (b) the absolute count. Comparison is made between the sepsis cohort and the two control groups.
Time frame: Between 48 and 72 hours after enrollment (preferably day 3)
Percentage and Absolute Count of Monocytes in Bone Marrow
Monocytes are defined as CD45⁺ CD3- CD56- HLA-DR⁺ CD14⁺ CD15- cells. The outcome is measured as both (a) the percentage among total bone marrow nucleated cells, and (b) the absolute count. Comparison is made between the sepsis cohort and the two control groups.
Time frame: Between 48 and 72 hours after enrollment (preferably day 3)
Percentage and Absolute Count of Neutrophils in Bone Marrow
Neutrophils are defined as CD45⁺ CD3- CD56- CD11b⁺ CD14- CD15⁺ cells. The outcome is measured as both (a) the percentage among total bone marrow nucleated cells, and (b) the absolute count. Comparison is made between the sepsis cohort and the two control groups.
Time frame: Between 48 and 72 hours after enrollment (preferably day 3)
Correlation Between Bone Marrow GMP-to-CLP Ratio (Percentage and Absolute) and SOFA score
Correlation between the baseline bone marrow GMP-to-CLP composite indices (both ratio based on percentages and ratio based on absolute counts) and SOFA score. SOFA score is calculated at enrollment and on days 1, 3, and 5. Correlation analyzed using Spearman correlation coefficients.
Time frame: Baseline (bone marrow at 48-72 hours) and days 1, 3, 5, 7 (severity scores)
Correlation Between Bone Marrow GMP-to-CLP Ratio (Percentage and Absolute) and APACHE II score
Correlation between the baseline bone marrow GMP-to-CLP composite indices (both ratio based on percentages and ratio based on absolute counts) and APACHE II score. APACHE II score is calculated at enrollment and on days 1, 3, and 5. Correlation analyzed using Spearman correlation coefficients.
Time frame: Baseline (bone marrow at 48-72 hours) and days 1, 3, 5, 7 (severity scores)
Association Between Bone Marrow Myeloid Lineage Bias (Percentages and Absolute Counts) and 90-Day Secondary Infection Rate
Evaluate whether the degree of baseline bone marrow myeloid lineage bias (measured by GMP-to-CLP ratio, percentage and absolute counts of PMN-MDSCs and M-MDSCs, and absolute count of PD-L1⁺ MDSCs) is independently associated with the occurrence of secondary infections within 90 days. Secondary infection is defined as a new infectious episode occurring ≥48 hours after enrollment, diagnosed by positive culture and/or clinical criteria requiring new antibiotic therapy. Logistic regression will be used, adjusting for baseline SOFA and APACHE II scores.
Time frame: From enrollment through 90-day follow-up
Bone Marrow Myeloid Lineage Bias (Percentages and Absolute Counts) as Predictors of 90-Day All-Cause Mortality
Evaluate the predictive value of baseline bone marrow cellular indices (GMP-to-CLP ratio, percentage and absolute counts of PMN-MDSCs and M-MDSCs, and absolute count of PD-L1⁺ MDSCs) for 90-day all-cause mortality. Survival analysis using Kaplan-Meier curves with Log-rank tests, and Cox proportional hazards regression calculating hazard ratios adjusted for age, SOFA score, and APACHE II score.
Time frame: From enrollment through 90-day follow-up
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.
Association Between Bone Marrow Myeloid Lineage Bias (Percentages and Absolute Counts) and ICU / In-Hospital Mortality
Evaluate the association between baseline bone marrow myeloid lineage bias indices (percentage and absolute counts of progenitor and MDSC subsets) and short-term mortality endpoints, including ICU mortality and in-hospital mortality. Logistic regression will be used to calculate odds ratios adjusted for baseline severity scores.
Time frame: From enrollment through hospital discharge (up to 90 days)
Correlation Between Bone Marrow MDSC Absolute Counts and Peripheral Blood Lymphocyte Absolute Counts
Assess the correlation between bone marrow PMN-MDSC and M-MDSC frequencies and absolute counts (along with their PD-L1 expression) and peripheral blood lymphocyte subset absolute counts per μL (and percentages). Peripheral blood immunophenotyping includes: CD3⁺ total T cells, CD3⁺CD4⁺ helper T cells, CD3⁺CD8⁺ cytotoxic T cells, CD3-CD16⁺CD56⁺ NK cells, and CD19⁺ B lymphocytes. This provides mechanistic evidence linking the absolute burden of bone marrow MDSCs to the severity of systemic lymphopenia-a hallmark of sepsis-induced immunosuppression.
Time frame: Baseline (bone marrow and peripheral blood at 48-72 hours; peripheral blood also on days 1, 5, 7 for longitudinal correlation)
Association Between Bone Marrow Myeloid Progenitor Absolute Counts and Peripheral Blood Cytokine Profile
Association between baseline bone marrow absolute counts of myeloid progenitors (GMPs, total MDSCs) and peripheral blood serum cytokine levels (measured in pg/mL). Serum cytokines are measured by multiplex assay and include: IL-2, IL-4, IL-6, IL-10, IL-17A, IFN-γ, and TNF-α. Inflammatory markers CRP and PCT are measured on days 1, 3, 5, and 7.
Time frame: Baseline (bone marrow and peripheral blood at 48-72 hours); CRP/PCT on days 1, 3, 5, 7
Longitudinal Dynamics of Peripheral Blood Lymphocyte and MDSC Percentages and Absolute Counts During the First Week
Track the dynamic changes in peripheral blood percentages and absolute counts per μL of CD3⁺ T cells (and CD4⁺/CD8⁺ subsets), CD19⁺ B cells, CD3-CD16⁺CD56⁺ NK cells, PMN-MDSCs, and M-MDSCs on days 1, 3, 5, and 7 after enrollment. Day 1 is the early systemic immune baseline preceding peak bone marrow remodeling (day 3); day 3 coincides with the primary aspiration for direct cross-sectional correlation; days 5 and 7 track post-remodeling trajectories. This temporal hierarchy tests whether bone marrow absolute myeloid burden drives subsequent peripheral immunosuppression.
Time frame: Days 1, 3, 5, 7 post-enrollment (peripheral blood)
Bone Marrow Microenvironment Cytokine and Chemokine Profile (Concentration)
Cytokine and chemokine concentrations (pg/mL) in bone marrow supernatant obtained from the aspirate, measured by ELISA or Luminex multiplex assay. Analytes include: G-CSF, GM-CSF, IL-6, IL-1β, TNF-α, IL-10, CXCL1, CXCL12, and CCL2. Levels are compared across the three cohorts to identify sepsis-specific bone marrow microenvironment alterations that may drive the myeloid lineage bias and MDSC accumulation.
Time frame: Baseline (bone marrow supernatant at 48-72 hours)
Exploratory Single-Cell Transcriptomic Profiling of Bone Marrow Hematopoietic Cells
Single-cell RNA sequencing (scRNA-seq) on bone marrow mononuclear cells from a subset of sepsis patients and healthy controls to identify differentially expressed genes and transcriptional pathways associated with myeloid lineage bias. Includes differential expression analysis and pathway enrichment (e.g., IFN signaling, STAT3, CEBPB, emergency granulopoiesis signatures).
Time frame: Baseline (bone marrow aspirate at 48-72 hours)
Correlation Between Gut Microbiota Alpha Diversity and Bone Marrow Myeloid Lineage Bias
Assess the correlation between gut microbiota alpha diversity indices (Shannon index, Simpson index, Chao1 index) and bone marrow myeloid lineage bias indices, including: GMP-to-CLP ratio, percentage and absolute counts of PMN-MDSCs and M-MDSCs, PD-L1 expression on MDSCs, and HSC/CMP/MEP/CLP counts.
Time frame: At 48-72 hours post-enrollment (rectal swab and bone marrow synchronized)
Correlation Between Key Dysbiotic Gut Bacterial Taxa and Bone Marrow Myeloid Lineage Bias
Identify key gut bacterial taxa that are significantly altered in sepsis and assess their correlation with bone marrow myeloid lineage bias indices, including: GMP-to-CLP ratio, percentage and absolute counts of PMN-MDSCs and M-MDSCs, PD-L1 expression on MDSCs, and HSC/CMP/MEP/CLP counts.
Time frame: At 48-72 hours post-enrollment (rectal swab and bone marrow synchronized)
Association Between Gut Microbiota-Derived Metabolites and Bone Marrow Myeloid Lineage Bias
Identify gut microbiota-derived metabolites associated with bone marrow myeloid lineage bias using untargeted metabolomics (LC-MS/MS) on rectal swab samples. Metabolite profiles are correlated with bone marrow myeloid lineage bias indices, including: GMP-to-CLP ratio, percentage and absolute counts of PMN-MDSCs and M-MDSCs, PD-L1 expression on MDSCs, and HSC/CMP/MEP/CLP counts.
Time frame: Rectal swab and bone marrow at 48-72 hours post-enrollment (synchronized)