Introduction
Severe burn injuries trigger a profound systemic response that extends far beyond local tissue damage. The magnitude of this response is largely determined by the extent of skin involvement, traditionally assessed by the depth of injury and the percentage of total body surface area (TBSA) affected. Although most burn injuries are nonfatal, extensive burns are associated with substantial morbidity and mortality worldwide.1,2
In the acute phase, major burns represent a prototypical model of systemic inflammatory response syndrome (SIRS). The release of inflammatory mediators increases capillary permeability, promotes fluid shifts, and necessitates prompt and careful fluid resuscitation to prevent hypovolemic shock and microcirculatory impairment.3,4 This early hyperinflammatory state is accompanied by a progressive disruption of immune regulation, predisposing patients to secondary organ dysfunction and infectious complications.
Given this dual pattern of initial hyperinflammation followed by relative immunosuppression, there is growing interest in therapeutic strategies aimed at restoring immune balance rather than simply suppressing inflammation. Among extracorporeal blood purification techniques, hemoadsorption has emerged as a potential immunomodulatory approach for critically ill patients, including those with severe burns.5 Several devices have been developed to attenuate circulating inflammatory mediators and modulate immune activation in intensive care settings.6
Previous reviews have addressed hemoadsorption broadly across heterogeneous critically ill or septic populations. In contrast, the present review is centered on the burn-specific context and explicitly separates evidence obtained in burn patients from data extrapolated from other conditions. The aim of this narrative review is to summarize the cellular and pathophysiological rationale for hemoadsorption in burn-induced immune dysregulation, to critically appraise the available preclinical and clinical evidence and its methodological limitations, and to propose candidate indications, treatment timing, safety considerations, and research priorities to guide its future evaluation.
Pathophysiology of burn-induced inflammation and immune dysregulation
Local tissue injury and regulated cell death
Burn wounds are classically described according to the three concentric zones defined by Jackson in 1947: the zone of coagulation, zone of stasis, and zone of hyperemia.7 The central zone of coagulation corresponds to irreversibly damaged tissue that typically evolves toward necrosis within the first 48 h. Surrounding this core is the zone of stasis, characterized by impaired microvascular perfusion and intense inflammation, while the outermost zone of hyperemia retains largely preserved perfusion and is potentially salvageable in the absence of secondary insults.8
Under steady-state conditions, dying cells are efficiently cleared by phagocytes, thereby limiting inflammatory signaling. However, in extensive burns, the abrupt and massive release of cellular debris overwhelms the local clearance mechanisms. Consequently, intracellular components are liberated into the extracellular space, acting as damage-associated molecular patterns (DAMPs) that amplify innate immune activation.
After a burn injury, thermal stress is associated with immediate cellular loss in the zone of coagulation and delayed cellular loss in the zone of stasis. In the zone of necrosis, the main mechanism of cellular loss is accidental cell death, which results in plasma membrane disruption.9 In the zone of stasis, sublethal heat leads to cell destruction through regulated cell death (RCD).10 Several types of RCD have been described, including apoptosis, necroptosis, ferroptosis, NETotic cell death, and autophagy-dependent cell death. During burn injury, several RCD processes occur, including apoptosis, necroptosis, and anoikis.11 The best-known form of regulated cell death is apoptosis, a term coined in 1972 to describe a morphological change in cells that is distinct from necrosis.12
Apoptosis occurs in various pathological and physiological processes and is characterized by morphological changes, such as cell rounding, nuclear condensation, membrane blebbing, and apoptotic body formation.13 Apoptosis follows two major pathways: intrinsic and extrinsic. Intrinsic apoptosis is mediated by mitochondrial permeabilization, whereas extrinsic apoptosis is mediated by cell membrane receptors. Both lead to the activation of caspases, which are involved in the controlled destruction of cellular components. Activation of inflammatory caspases can promote inflammation.14 Several pathways may induce intrinsic apoptosis in response to burns.11 Heat-sensitive transient receptor potential vanilloid channels can open at high temperatures, increasing intracellular calcium levels and leading to mitochondrial dysfunction.15 Thermal stress can induce a direct caspase-dependent pathway, and hyperthermia engages the endoplasmic reticulum in a stress response, which leads to caspase cleavage.16-18 However, in vivo studies of burn injury are limited, precluding precise identification of all underlying apoptotic pathways.
Necroptosis is an RCD that was first described in a burn model in 2015.19 Execution of necroptosis requires assembly of the necrosome, which is composed of receptor-interacting protein kinases 1 and 3 (RIPK1 and RIPK3),20 following activation of death receptors or innate immune sensors, including tumor necrosis factor receptor 1, Toll-like receptors, or the DNA-dependent activator of interferon regulatory factors.21 In contrast to apoptosis, in which release of proinflammatory components is limited, necroptosis can promote inflammation by releasing DAMPs and activating the inflammasome through RIPK3, thereby increasing IL-1β production.22
Anoikis is a form of intrinsic apoptosis that can be initiated by loss of integrin-mediated cellular attachment to the extracellular matrix.23 This form of apoptosis has been described in keratinocytes.24 Heat shock can induce integrin downregulation and cell shedding.25 This phenomenon may occur in burn wounds and may be enhanced by degradation of the extracellular matrix by metalloproteinases, the expression of which may be induced by heat shock.26 Although the rate of apoptosis in the skin of patients with burns is higher than that in healthy individuals, the contribution of anoikis remains unclear.27
From local inflammation to systemic response
When the burn size exceeds approximately 20–30% TBSA, local inflammatory signals propagate systemically, engaging hormonal, metabolic, and immune pathways. The accumulation of cellular debris and DAMPs surpasses the clearance capacity and can precipitate SIRS. This response unfolds in two overlapping phases: an early hyperinflammatory state, followed by a compensatory anti-inflammatory phase that may culminate in immune suppression.28
During the proinflammatory phase, both pro- and anti-inflammatory cytokine levels are markedly elevated.29 Their expression is largely governed by p38 MAPK and NF-κB signaling pathways, which drive transcription of TNF-α, IL-1β, and IL-6.30,31 IL-6 levels are typically elevated after burn injury and peak within 4 days, although the timing may vary with burn extent.32,33 DAMPs further prime neutrophils and macrophages, intensifying local and systemic inflammatory responses. Dysregulated macrophage activity after burn injury appears to be a key contributor to systemic immune dysregulation in patients with burns.34 Macrophage polarization toward a classically activated phenotype enhances production of IL-1, IL-6, and TNF-α.35 Macrophages are also major producers of inducible nitric oxide synthase,36 which promotes vasodilation and perpetuates NF-κB activation, establishing a self-amplifying inflammatory feedback loop.37 These processes contribute to immune cell activation, increased vascular permeability, coagulation activation, and SIRS.
Compensatory anti-inflammatory response
As in sepsis, the initial hyperinflammatory phase is followed by a compensatory anti-inflammatory response.38 This phase is characterized by widespread immune attenuation, including reduced lymphocyte responsiveness, impaired antigen presentation, and diminished dendritic cell function. Key mediators of this shift include TGF-β, IL-4, and especially IL-10, which exerts potent immunosuppressive effects on macrophages.39 At the cellular level, changes include lymphocyte anergy mediated by type 2 helper T cells,40 downregulation of monocyte function with reduced cytokine secretion and decreased human leukocyte antigen-DR (HLA-DR) surface expression,41 and a reduction in functional dendritic cells.42 The timing and magnitude of immune reprogramming vary among patients and may increase susceptibility to secondary infections and sepsis. This complex immune-inflammatory response is illustrated in Figure 1.
Immunomodulation approaches in burn care
Immune monitoring and specific therapies for modulating immunity in burn injuries remain limited. Most studies have been conducted in animal models, and data from patients with burns are scarce. Several treatments have been proposed to prevent immune dysregulation after burn injury, as described below.
Some authors have investigated the potential benefits of diets that may modulate the immune response. Diets supplemented with immunomodulatory nutrients such as n-3 fatty acids, glutamine, arginine, or antioxidants such as oral vitamins C and E have shown no clear effect on mortality, length of stay, or rates of burn wound or non-wound infection, although early small studies suggested that glutamine might reduce length of stay and mortality.43
Topical immunomodulation has been proposed to treat burn injuries without systemic effects, to minimize the progression of tissue injury, attenuate the local inflammatory response, or promote wound healing. For example, in rat models, administration of anti-TNF-α conjugates reduced the extent of necrosis,44 and administration of a p38 MAPK inhibitor reduced dermal apoptosis and attenuated burn-induced cellular stress.45 Topical administration of heat-killed bacteria, such as Lactobacillus, appears to promote wound healing in animal studies.46
Few studies have investigated immunostimulatory therapies with interferon gamma (IFN-γ) or granulocyte-macrophage colony-stimulating factor (GM-CSF).47 IFN-γ is a cytokine mainly produced by lymphocytes. It activates circulating monocytes and restores monocytic HLA-DR expression, thereby increasing their antigen-presenting capacity and phagocytic function.48 IFN-γ has been described as an immunomodulatory adjunctive therapy for invasive mucormycosis after severe burns.49 Data on GM-CSF use in patients with burns remain limited. Administration of this cytokine increases circulating granulocyte counts and modulates leukocyte function. It has been associated with increased superoxide production and reduced myeloperoxidase activity, but the overall effect on morbidity and mortality cannot be evaluated.50 Granulocyte colony-stimulating factor has also been studied in relation to survival in burn-induced sepsis.51 GM-CSF may be beneficial for burn wound healing.52
Hemoadsorption
Rationale for using hemoadsorption in patients with burns
Extensive burn injuries are characterized by the sustained release of inflammatory mediators that propagate beyond the wound and contribute to systemic organ dysfunction. In this context, extracorporeal blood purification techniques have been proposed as adjunctive tools to mitigate the inflammatory burden rather than solely targeting individual cytokines. Hemoadsorption has emerged as a strategy to “rebalance” the immune response by removing circulating mediators and modulating innate immune cell activity.
From a mechanistic perspective, the biological effects of hemoadsorption likely extend beyond simple plasma clearance. Because cytokines continuously equilibrate between the interstitial and intravascular compartments, adsorption within the extracorporeal circuit may gradually lower tissue mediator concentrations even when plasma levels appear only modestly reduced. This compartmental redistribution may attenuate downstream inflammatory signaling before critical activation thresholds are reached, thereby limiting the propagation of the pro-inflammatory cascade.53,54
At the cellular level, hemoadsorption has been suggested to exert indirect effects on immune cell function. Di Carlo’s “mediator delivery hypothesis,” originally proposed for hemofiltration, suggests that replacement-fluid infusion enhances lymphatic flow and mobilizes inflammatory mediators for clearance.55 Related extracorporeal approaches have also been reported to modify leukocyte phenotype and function, including the upregulation of HLA-DR expression on monocytes and selective reduction of activated neutrophils.56,57In vitro data additionally suggest a potential anti-apoptotic effect on tubular epithelial cells exposed to septic plasma, supporting a broader immunomodulatory impact beyond cytokine removal.58
In clinical practice, hemoadsorption can be applied using different configurations depending on the patient’s condition. In patients with burns who develop acute kidney injury, the cartridge is commonly integrated in series with continuous renal replacement therapy (CRRT), allowing simultaneous renal support and mediator removal. Alternatively, hemoadsorption may be used as a standalone extracorporeal therapy in selected patients with predominantly inflammatory instability.
Building on these considerations, several burn subpopulations can be proposed as plausible candidates for hemoadsorption. First are patients with extensive burns, typically involving more than 20 to 30% of TBSA, who develop an early and predominantly hyperinflammatory phenotype with vasoplegia and escalating vasopressor requirements; in these patients, the aim would be to attenuate the early cytokine surge that drives distributive shock. Second are patients with burn-related septic shock, in whom a secondary infection amplifies the systemic inflammatory response and for whom mediator removal might serve as an adjunct to source control and appropriate antimicrobial therapy. Third are patients who already require CRRT for acute kidney injury, in whom hemoadsorption can be delivered in series without the need for additional vascular access. With respect to timing, two distinct therapeutic windows can be distinguished. The first is an early hyperinflammatory window, within the first hours to days after injury, in which the rationale is to blunt the cytokine storm and limit remote organ injury. The second is a later window, dominated by infection and sepsis, in which the goal is to control the inflammatory response to secondary insults. These proposed indications and timing strategies remain hypothesis-generating and require prospective evaluation before they can be recommended in routine practice.
Modern hemoadsorption relies predominantly on biocompatible polystyrene-divinylbenzene sorbents composed of coated macroporous beads. Among the currently available devices, CytoSorb® (CytoSorbents Corporation, Princeton, NJ, USA) and HA-series cartridges (Jafron Biomedical, Zhuhai, China) are the most widely used.59 These materials can adsorb a broad range of hydrophobic and protein-bound molecules, including complement components and multiple cytokines.60 The clinical significance of reducing circulating pro- and anti-inflammatory mediators with extracorporeal blood purification remains uncertain.53,61 Although current studies suggest possible improvements in parameters such as hemodynamics and duration of mechanical ventilation, the effect of hemoadsorption on mortality remains uncertain.62
Results from trials assessing hemoadsorption in burns
Evidence supporting the clinical use of modern hemoadsorption sorbents in patients with burns remains scarce and largely exploratory. Most available data are derived from small case series, retrospective analyses, or animal models, which limit the generalizability of the findings.
Experimental work in a porcine burn model demonstrated that CytoSorb® cartridges effectively reduced cytokine concentrations across the device; however, this did not translate into a significant decrease in circulating plasma IL-1, IL-6, or IL-10 levels.63 Conversely, studies in healthy volunteers exposed to lipopolysaccharide have shown that hemoadsorption can lower the systemic concentrations of TNF, IL-6, and IL-8, suggesting that efficacy may depend on the underlying inflammatory context.64 The studies summarized here do not isolate the clinical effect of hemoadsorption in an adequately powered randomized trial. The clinical evidence summarized here includes a retrospective cohort, isolated case reports, and a small randomized study of combined hemofiltration and hemoperfusion, which are detailed below and summarized in Table 1.63-68
| Study (reference) | Design and population | Main findings | Main limitations |
|---|
| Preclinical study63 | Experimental porcine burn model (animal) | CytoSorb reduced cytokine concentrations across the cartridge, but without a significant fall in circulating IL-1, IL-6, or IL-10 | Animal model; no clinical outcomes; cartridge effect not reflected in plasma levels |
| Endotoxin challenge64 | Healthy human volunteers exposed to lipopolysaccharide (non-burn) | Reduction in systemic TNF, IL-6, and IL-8, suggesting context-dependent efficacy | Non-burn experimental model; healthy subjects; surrogate biological endpoints only |
| Mariano et al.65 | Single-center retrospective cohort, 35 patients with burns (CytoSorb + CRRT, n = 11, vs CRRT alone, n = 24) | Improved hemodynamic stability and reduced norepinephrine requirements in the CytoSorb group | Small sample; single-center; nonrandomized; likely confounding by indication; no established causal effect on mortality or organ function |
| Case reports66,67 | A conference abstract and a case report in patients with burns | Demonstrated feasibility; one report described a reduction in bilirubin after hemoadsorption | Anecdotal; no control group; no robust clinical endpoints; high risk of reporting bias |
| Randomized study68 | Randomized study in 40 patients with burns (20 per group; China); hemofiltration plus hemoperfusion versus conventional care | Suggested improvement in biochemical markers and septic complications | Small sample; combined intervention prevents attribution to hemoadsorption alone; sorbent incompletely specified in the English abstract |
Clinical data on patients with burns are extremely limited. A recent single-center retrospective study including 35 patients (11 receiving CytoSorb plus CRRT and 24 receiving CRRT alone) reported improved hemodynamic stability and reduced norepinephrine requirements when CytoSorb® was combined with CRRT compared with CRRT alone. However, this study was limited by its small sample size, single-center design, and lack of randomization.65 Because treatment was not allocated at random, this observational comparison is particularly susceptible to confounding by indication, and the reported survival advantage should be regarded as hypothesis-generating rather than confirmatory. An additional limitation concerns the nonrandomized comparison: the incremental benefit specifically attributable to the adsorptive cartridge cannot be reliably inferred from this observational comparison with CRRT alone and therefore remains uncertain. This caveat is consistent with the broader critical care literature: although a recent meta-analysis restricted to septic shock suggested a possible short-term survival and hemodynamic benefit of hemoadsorption,69 randomized data across mixed indications have shown no consistent mortality reduction, and one meta-analysis restricted to randomized controlled trials reported a signal toward higher mortality.70,71
These observations underscore a persistent gap between the hemodynamic and biochemical improvements attributed to hemoadsorption and patient-centered outcomes; no definitive benefits in mortality or ventilator-free days have been demonstrated in patients with burns.72 This cautious interpretation is consistent with a recent international consensus, which regards contemporary hemoadsorption as a novel and experimental intervention supported by limited data and a broad research agenda,73 and with critical appraisals concluding that no study has yet demonstrated a survival benefit and that the current evidence does not justify its routine use.74
Additional reports are primarily limited to isolated case reports. These reports provide feasibility data but lack robust clinical endpoints. One case report described a reduction in bilirubin levels after hemoadsorption, whereas another failed to present interpretable outcome data despite the initial objectives.66,67 A randomized Chinese study of 40 patients suggested potential improvements in biochemical markers and septic complications with intermittent hemofiltration combined with hemoperfusion; however, the combined intervention and incompletely specified sorbent in the English abstract limit attribution to hemoadsorption alone.68
Overall, existing studies suggest that hemoadsorption is technically feasible and may influence hemodynamic or biological parameters; however, they do not provide high-quality evidence for improved survival or organ function in patients with burns.
Technical considerations
The technical parameters, such as the optimal timing of initiation and duration of hemoadsorption, remain unknown. The timing of the intervention is particularly challenging in patients with burns. For example, early initiation aims to mitigate the “cytokine storm,” SIRS-related vasoplegia, and remote organ injury and dysfunction.75,76 Conversely, the use of extracorporeal blood purification as a “late” adjuvant therapy when renal replacement therapy is needed for renal support has also been evaluated. Chung et al. evaluated hemofiltration in adults with severe burns and acute kidney injury in cohort studies and a randomized controlled trial.77-79 It should be emphasized, however, that these data were generated with hemofiltration, a convective clearance technique that is mechanistically distinct from adsorption. Such evidence cannot be extrapolated to support hemoadsorption, and the two modalities should not be regarded as interchangeable; this distinction is maintained throughout the present review. Other available hemoadsorption devices have been only sparsely evaluated in patients with burns, and robust controlled evidence remains lacking. These devices can be categorized into pure sorbents, which may exhibit selective or non-selective adsorption, and hemofilters with hemoadsorption capacity, which can also vary in selectivity.
Among non-selective sorbents, polystyrene resin sorbents are designed for the adsorption of medium- to high-molecular-weight metabolites and protein-bound uremic retention solutes.80 Among selective sorbents, some devices are specifically designed for endotoxin removal in the septic population, such as the polymyxin B-immobilized cartridge (Toraymyxin®) and synthetic cationic peptide-based lipopolysaccharide adsorbers.81 Among selective hemoadsorption devices, heparin-grafted polyethylene beads are designed to mimic the endothelial glycocalyx and remove a variety of pathogens, including bacteria, viruses, and fungi, from the bloodstream. They may warrant evaluation in patients with burns because of the high incidence and severity of bloodstream infections in this population.82 The PURIFY-RCT trial has been registered (NCT05011656).83
Some hemofilters have adsorptive properties, including nonselective adsorptive membranes such as surface-treated acrylonitrile 69 (AN69ST), polymethyl methacrylate (PMMA), and oXiris membranes. They may be considered in septic shock associated with acute kidney injury requiring CRRT.84 Another technique is the selective cytopheretic device (SCD; SeaStar Medical, Inc.). SCD is an extracorporeal synthetic membrane device designed to reversibly bind activated circulating neutrophils and monocytes in a low-ionized-calcium environment and under low-shear blood flow.85,86 A multicenter, randomized, controlled trial (NCT05758077) is ongoing to assess the efficacy and safety of SCD in patients with acute kidney injury requiring CRRT.87,88 Despite advances in sorbent technology, no robust study has provided conclusive evidence supporting the clinical use of hemoadsorption, particularly in patients with burns. Further animal and human studies are needed to characterize the performance of each hemoadsorption cartridge.89
A comparison of the main extracorporeal blood purification devices, their mechanisms, candidate clinical scenarios, and the level of evidence in burns is summarized in Table 2.
| Device (sorbent) | Mechanism and selectivity | Potential clinical scenario | Evidence in burns |
|---|
| CytoSorb (polystyrene-divinylbenzene) | Non-selective adsorption of mid-molecular-weight molecules (approximately 5-60 kDa), including cytokines | Hyperinflammation and cytokine storm, septic shock; commonly used in series with CRRT | Direct burn data available (porcine model, retrospective cohort, case reports); no controlled study demonstrating an outcome benefit |
| Jafron resin cartridges (e.g., HA330/HA380) | Non-selective resin adsorption of cytokines and middle molecules | Cytokine storm and sepsis; standalone or combined with CRRT | Limited to case-level reports in patients with burns; no controlled study (literature search updated September 2026)
|
| Polymyxin B-immobilized fiber (e.g., Toraymyxin/PMX) | Selective adsorption of endotoxin (lipopolysaccharide) | Gram-negative sepsis and endotoxemia | No study in patients with burns identified (literature search updated September 2026)
|
| Seraph 100 (heparin-grafted) | Selective binding of circulating pathogens and pathogen-associated molecules | Bloodstream infection and pathogen reduction | No controlled study in patients with burns identified (literature search updated September 2026)
|
| Adsorptive membranes (AN69ST, oXiris, PMMA) | Hemofilters that combine convective/diffusive clearance with surface adsorption; selectivity varies | Combined renal support and mediator removal in acute kidney injury | Limited to a small case series of patients with burns treated with oXiris, reporting improved urine output and SOFA score; no controlled study. Mechanistically distinct from pure adsorption and should not be conflated with it (literature search updated September 2026)
|
| Selective cytopheretic device (SCD) | Immunomodulation by reversible binding of activated neutrophils and monocytes in a low ionized-calcium, low-shear environment | Acute kidney injury requiring CRRT; hyperinflammatory states | No study in patients with burns identified; a multicenter randomized controlled trial is ongoing in AKI requiring CRRT (NEUTRALIZE-AKI, NCT05758077) (literature search updated September 2026) |
Safety, complications, and contraindications
An additional concern is the non-selective nature of many sorbents. In addition to harmful inflammatory mediators, potentially beneficial molecules may also be removed from the circulation. Moreover, correction of treatment-related imbalances through transfusion of blood products or albumin could paradoxically trigger secondary inflammatory responses, counteracting some of the intended immunomodulatory effects.
Current experience indicates that polystyrene-divinylbenzene hemoadsorption cartridges are generally well tolerated, with moderate and transient reductions in platelet count and serum albumin being the most consistently reported laboratory changes.90 Nevertheless, hemoadsorption remains an extracorporeal therapy that requires vascular access and an individualized anticoagulation strategy. As such, it carries the inherent risks of bleeding, thrombosis, and catheter-related infection, comparable to other renal replacement or extracorporeal techniques.91-93 Additionally, the correction of hemoadsorption-induced imbalances through blood, platelet, or albumin transfusion may lead to immunological complications and potentially counteract the intended immunomodulatory effects of hemoadsorption by inducing a pro-inflammatory immune response.94,95 Particular attention should be paid to the consequences of the non-selective nature of adsorption, which is not restricted to pro-inflammatory cytokines. The removal of albumin reduces oncotic pressure and, because many drugs are highly protein-bound, may alter the free fraction and clearance of co-administered agents. Coagulation factors may likewise be adsorbed, which, together with the fall in platelet count described above, can aggravate the bleeding tendency already present in critically ill patients with burns. Of particular concern is the clearance of anti-infective agents: hemoadsorption can substantially remove several antibiotics and antifungals, potentially producing subtherapeutic plasma concentrations and treatment failure at a time when severe infection is a leading cause of death in burns, and in a population in which antimicrobial pharmacokinetics are already profoundly altered by hypermetabolism, capillary leak, and augmented renal clearance. Therapeutic drug monitoring and device-specific dose adjustment, without delaying clinically indicated anti-infective treatment, are therefore advisable. Finally, because the cartridge does not discriminate between mediators, beneficial anti-inflammatory cytokines such as interleukin-10 may be removed alongside pro-inflammatory ones, raising the theoretical risk of blunting the compensatory anti-inflammatory response and impairing host defense.72
No universally accepted absolute contraindications to hemoadsorption have been established to date, although potential complications and patient-specific clinical and technical contraindications should be considered.73 Active uncontrolled bleeding or severe refractory coagulopathy constitutes a relative contraindication, given the additional loss of platelets and coagulation factors and the potential need for anticoagulation to maintain the extracorporeal circuit. The absence of safe vascular access and profound hemodynamic instability precluding the maintenance of adequate extracorporeal blood flow are further practical limitations. Concomitant treatment with anti-infective agents with a narrow therapeutic index calls for close monitoring rather than precluding therapy. In all cases, the decision to initiate hemoadsorption in a patient with burns should rest on an individualized risk–benefit assessment, with explicit attention to coagulation status, vascular access, hemodynamic reserve, and ongoing drug therapy.72
Taken together, these considerations highlight the need for individualized patient selection, careful timing of therapy, and close biological monitoring when hemoadsorption is used in patients with severe burns.
Future investigations in the sorbent field
Future research must move beyond descriptive and observational studies toward adequately powered randomized controlled trials focused on clinically meaningful endpoints such as organ failure burden, ventilator-free days, and mortality. However, several fundamental questions remain unanswered. These include the extraction ratios of different cartridges, total mass of mediators removed over time, and the kinetics of sorbent saturation. These parameters likely vary across devices and clinical contexts, emphasizing the need for device-specific evaluations rather than extrapolations from heterogeneous studies. Determining the optimal timing of initiation is critical. Early application may be most appropriate to evaluate in patients with extensive burns and marked hyperinflammation, whereas delayed use may be more appropriate to evaluate in those who develop acute kidney injury requiring renal replacement therapy.
Finally, the potential role of hemoadsorption as a standalone therapy versus its combination with CRRT or other immunomodulatory strategies warrants systematic investigation in well-designed comparative trials.
Strengths and limitations
A strength of this review is its burn-specific focus and its explicit distinction between direct evidence from patients with burns and evidence extrapolated from other critically ill populations. Several limitations of the present review and of the underlying evidence base should also be acknowledged, and they temper all of the interpretations proposed above. First, the studies summarized here do not isolate the clinical effect of hemoadsorption in an adequately powered randomized trial. The clinical evidence summarized here includes a retrospective cohort, isolated case reports, and a small randomized study of combined hemofiltration and hemoperfusion. These data preclude firm causal attribution to hemoadsorption alone, and the absence of randomization in most reports renders the reported benefits particularly susceptible to confounding by indication. Second, the available studies are small, mostly single-center, and methodologically heterogeneous with respect to the device used, the timing and duration of therapy, the dose delivered, and the concomitant use of renal replacement therapy. This heterogeneity precludes any quantitative synthesis and limits comparability across reports. The nonrandomized comparison and potential differences in concomitant CRRT delivery are additional constraints on attributing any incremental effect to adsorption itself. Third, most reported endpoints are intermediate rather than patient-centered. Improvements in vasopressor requirements, hemodynamic indices, or circulating mediator concentrations have not translated into definitive evidence of benefits in mortality, organ failure burden, or ventilator-free days in patients with burns. Preclinical work illustrates the same gap, as cartridge-level clearance has not consistently been reflected in systemic cytokine concentrations. Fourth, a substantial part of the supporting rationale is extrapolated from non-burn critically ill and septic populations. Because burn pathophysiology, pharmacokinetics, and the time course of immune dysregulation differ substantially from those of other critically ill patients, such extrapolation is uncertain. For the same reason, evidence generated with mechanistically distinct modalities, in particular hemofiltration, cannot be used to support hemoadsorption. Finally, safety data specific to patients with burns are scarce. The consequences of non-selective adsorption, notably the removal of albumin, coagulation factors, anti-infective agents, and anti-inflammatory cytokines, have not been quantified in this population, in which antimicrobial pharmacokinetics are already profoundly altered.