Dendritic-cell depletion and dysfunction in sepsis: a compartment- and evidence-aware synthesis.

Sepsis-associated immune dysfunction varies across patients, compartments, and sampling windows. Dendritic cells (DCs) offer a useful-but neither exclusive nor primary-lens because they connect innate sensing with antigen presentation, cytokine instruction, and adaptive immunity. Human studies show reduced circulating DC abundance, altered phenotype, checkpoint context, and impaired ex vivo responsiveness, whereas experimental studies demonstrate splenic DC loss, disturbed bone-marrow generation, impaired anigen-specific T-cell priming, and tissue-specific pDC vulnerability or rescue. These observations support four distinct, potentially coexisting contributions: depletion, dysfunction of residual cells, redistribution, and impaired replacement. A low blood count cannot distinguish among them. Across mechanisms, evidence for developmental supply, survival and cell death, trafficking, antigen presentation, cytokine output, regulatory-like programming, checkpoint signaling, and organelle or metabolic stress varies substantially in identity confidence, model, functional validation, and causal support. Accordingly, human blood evidence is not treated as equivalent to murine marrow, spleen, or lung perturbation, and transcriptomic APC/DC-like states are not equated with demonstrated cellular function. The mouse mregDC-like program remains a Class 4 exploratory regulatory-like state with substantial within-study support but no validated human sepsis homolog. Apparent conflicts-including early inflammatory amplification versus persistent immune suppression, and IL-3-associated inflammation versus late pDC-dependent antiviral rescue-are interpreted as context-, time-, and compartment-dependent rather than forced into a universal sequence. DC-informed monitoring should complement established broader approaches such as monocyte HLA-DR and transcriptomic immune subgrouping, using a transparent ten-domain appraisal of identity and evidence. No DC-only endotype, treatment trigger, bedside biomarker, or DC-directed therapy is validated. This synthesis defines testable, timing-aware, and safety-bounded research hypotheses while keeping clinical claims proportional to the evidence.