Gelsolin Levels Track with Sepsis Mortality, But Diagnostic Use Remains Limited
A systematic review and meta-analysis of circulating gelsolin in sepsis points to a consistent signal: patients who died had lower levels than survivors. Across seven studies reporting survival outcomes, the pooled ratio of means was 0.62, with a 95% confidence interval of 0.53 to 0.73, indicating a substantial drop in circulating gelsolin among non-survivors. The direction of effect held across subgroup analyses, yet the data were far from uniform. Heterogeneity was high, with an I² of 74%, which means the magnitude of the association varied materially from study to study rather than marching in lockstep.
One clinically relevant detail stands out. An admission plasma gelsolin threshold of 113.6 mg/L was linked to discrimination for 28-day mortality in a single study, but that cut-off remains exploratory. Assay variation, shifting sample handling methods, and differences in reported thresholds make any one number difficult to transplant into routine bedside use. In practice, gelsolin looks more like a risk marker in development than a ready-made triage tool.
Why Gelsolin Fits the Biology of Severe Infection
Gelsolin is not a random candidate biomarker chosen after the fact. It is an actin-binding protein with actin-severing, capping, and nucleating functions, which means it helps regulate cytoskeletal dynamics and cell behavior. In severe infection, widespread tissue injury releases actin into the circulation, and gelsolin participates in actin scavenging. That biological role gives the marker a plausible link to the scale of tissue damage and inflammatory burden unfolding during sepsis.
The inflammatory relevance is equally important. Gelsolin and related proteins have been tied to inflammatory signaling, including interactions with bacterial products and Toll-like receptor pathways. In a syndrome driven by dysregulated host response, endothelial injury, immune activation, and multiorgan stress, a decline in circulating gelsolin could reflect depletion of a protective buffer rather than a mere laboratory curiosity. Experimental work cited in the review suggests that exogenous gelsolin may attenuate injury in sepsis-related models, which makes the biomarker signal biologically coherent even if clinical deployment is still premature.
Diagnostic Performance Was Only Modest
Three studies contributed to the exploratory diagnostic analysis, and here the picture became less convincing. Pooled sensitivity reached 0.81, suggesting gelsolin identified a fair proportion of patients with sepsis. Specificity was only 0.61, leaving a substantial false-positive burden. The positive likelihood ratio was 2.19 and the diagnostic odds ratio was 6.38, both of which indicate some discriminatory value but not the kind of separation needed for stand-alone diagnosis.
The specificity problem is not a minor statistical footnote. Circulating gelsolin is also reduced in other inflammatory and tissue-injury conditions, so low values do not point uniquely to sepsis. Substantial heterogeneity in the diagnostic analysis, with I² at 90.6%, further erodes confidence in a single pooled estimate. Different comparator groups and different diagnostic settings likely contributed to the spread, but the practical consequence is the same: a low gelsolin result cannot reliably distinguish sepsis from other serious illnesses on its own.
The bedside implication is straightforward. Gelsolin may add signal to a broader assessment, but it does not replace blood cultures, clinical examination, organ dysfunction scoring, or established laboratory workup.
What the Review Included and How the Evidence Was Weighed
Investigators searched PubMed, Scopus, Web of Science, Embase, the Cochrane Library, and clinical trial registries through July 26, 2026. Eight studies met eligibility criteria. Seven were usable for prognostic comparison between survivors and non-survivors, and three contributed to diagnostic analysis. Eligible studies had to include adult patients with sepsis and report circulating gelsolin levels with extractable data for prognosis or diagnosis.
The review used standard methods for quality appraisal. Prognostic studies were evaluated with the Newcastle-Ottawa Scale, and diagnostic studies were assessed using QUADAS-2. Because the diagnostic evidence base was small, a stable bivariate or HSROC model could not be reliably estimated. That limitation matters because pooled diagnostic accuracy becomes fragile when only a few studies contribute data and the underlying methods differ widely.
Leave-one-out sensitivity analysis did not materially change the prognostic finding, which strengthens confidence that the mortality association was not driven by any single report. Even so, the overall evidence base remains small, and the studies included were not uniform in assay methods, comparator groups, or clinical context. Those differences help explain why the signal is compelling yet still unfinished.
A Biomarker with Prognostic Promise, Not Diagnostic Finality
The most defensible conclusion from the available evidence is narrow but useful: lower circulating gelsolin is associated with higher mortality risk in sepsis and may help with early prognostic risk stratification. That is a meaningful role in critical care, where identifying patients at greater risk can shape monitoring intensity, escalation decisions, and expectations about trajectory. But prognostic association is not the same as validated clinical utility.
Several obstacles remain before gelsolin can move closer to routine use. Standardized assays are still needed. Clinically relevant comparator groups are still needed. External validation of any proposed cut-off is still needed, especially because the 113.6 mg/L threshold came from one study and may not generalize across laboratories or patient populations. The review also makes clear that low gelsolin appears in other inflammatory and tissue-injury states, which limits its usefulness as a specific diagnostic marker for sepsis.
For now, gelsolin should be viewed as a biologically plausible candidate biomarker with a reproducible association to poor outcome, not as a standalone diagnostic answer. The field now needs larger multicenter prospective studies with harmonized measurement methods and clear clinical endpoints if gelsolin is to move from promising association to dependable practice tool.