Biosensors and Nanotheranostics

1. Introduction

Diabetes mellitus has become one of the defining public health problems of this century. It is a disorder of chronic hyperglycaemia, but its real burden lies in what sustained metabolic dysregulation does to nerves, vessels, and tissues over time. More than 537 million adults are currently living with diabetes, and that figure is projected to reach 783 million by 2045 (Renuka et al., 2023). Of the many complications that follow, few are as disabling, or as expensive as the diabetic foot ulcer (DFU).

The numbers are sobering. Between 19% and 34% of people with diabetes will develop a foot ulcer during their lifetime, and, because the underlying neurological and microvascular damage persists after closure, about 65% of healed ulcers recur within three to five years (Armstrong et al., 2017). The lifetime risk of lower-extremity amputation in this group approaches 20%. People with diabetes face a 10- to 20-fold higher risk of amputation than the general population, and an infected or non-healing ulcer precedes more than 80% of major amputations, with a corresponding rise in five-year mortality (Armstrong et al., 2017; Kushwaha et al., 2022).

Why do these wounds fail to heal? The answer is not a single defect but several that reinforce one another. Peripheral neuropathy removes the protective sensation that would normally warn a patient of repetitive plantar trauma, while autonomic dysfunction reduces sweating and leaves the skin dry and prone to fissuring, an easy entry point for microbes. Peripheral arterial disease and chronic ischaemia limit oxygen and nutrient delivery (Kushwaha et al., 2022; Renuka et al., 2023). At the cellular level, the normally ordered sequence of haemostasis, inflammation, proliferation, and remodelling breaks down. Persistent hyperglycaemia drives the accumulation of reactive oxygen species (ROS) and advanced glycation end-products (AGEs), and the wound becomes trapped in a self-perpetuating inflammatory phase (Renuka et al., 2023). Macrophages fail to switch from the pro-inflammatory M1 phenotype to the reparative M2 phenotype; interleukin-1α (IL-1α), interleukin-6 (IL-6), and tumour necrosis factor-α (TNF-α) remain elevated; matrix metalloproteinases (MMPs) degrade newly deposited extracellular matrix (ECM); and deficient signalling by vascular endothelial growth factor (VEGF), fibroblast growth factor (FGF), and platelet-derived growth factor (PDGF) leaves the wound bed poorly vascularised and slow to granulate (Khalid et al., 2026; Sajjad et al., 2025; Singla et al., 2018).

Infection compounds these problems. More than half of chronic diabetic ulcers show clinical signs of infection, usually polymicrobial, involving Gram-positive organisms such as Staphylococcus aureus and Streptococcus species together with Gram-negative pathogens such as Pseudomonas aeruginosa, Escherichia coli, and other Enterobacteriaceae (Kushwaha et al., 2022; Zheng et al., 2020). These organisms organise into dense biofilms encased in an extracellular polymeric substance (EPS) that shields them from host immune cells and from antibiotics, and that encourages multidrug-resistant phenotypes (Kart et al., 2024). When the microbial bioburden is not cleared, the consequences range from local necrosis to deep tissue invasion, osteomyelitis, and sepsis.

Standard care (surgical debridement, off-loading, systemic antimicrobials, and passive dressings such as gauze and absorbent pads) still leaves more than 30% of DFUs unhealed (Kushwaha et al., 2022). Conventional dressings are, for the most part, simple physical covers. They stick to the fragile wound bed, cause secondary trauma when removed, manage exudate poorly, and have no intrinsic antibacterial or regenerative function (Revin et al., 2023; Zheng et al., 2020). This mismatch between what the wound needs and what the dressing provides has pushed the field towards bio-interactive dressings that mimic the native ECM, keep the wound moist without macerating it, and actively suppress infection while supporting cell proliferation and angiogenesis (Gürsoy et al., 2025; Renuka et al., 2023).

Bacterial cellulose (BC) is one of the most promising candidates. Produced by aerobic bacteria such as Komagataeibacter (formerly Gluconacetobacter), Acetobacter, and Rhizobium, BC is secreted as an extracellular polysaccharide of linear β-1,4-glucan chains, free of the hemicellulose, pectin, and lignin that contaminate plant cellulose (Czaja et al., 2007; Revin et al., 2023). Its three-dimensional (3D) network of nanofibres, each 10–100 nm across, closely resembles the architecture of human dermis. This structure gives BC high wet tensile strength, a water-holding capacity of up to 99 wt.%, balanced vapour transmission, and good gas and nutrient permeability (Gürsoy et al., 2025; Zheng et al., 2020). Clinically, hydrated BC membranes act as a barrier to microbial entry, peel away without pain, support autolytic debridement, and maintain the moist environment that favours re-epithelialisation (Czaja et al., 2007; Yang, 2021).

There is one important limitation. Pristine BC has no intrinsic antimicrobial or antibiofilm activity, so in a heavily contaminated, exudative DFU it cannot, on its own, stop bacterial growth (Zheng et al., 2020). The response has been to build BC nanocomposites carrying nanoscale antimicrobial agents. The abundant surface hydroxyl groups, high surface area, and open porosity of the BC network make it a convenient template for the in situ or ex situ nucleation, deposition, and sustained release of nanoparticles (Ibraheem & Shaker, 2025; Revin et al., 2023).

Metallic and metal oxide nanoparticles dominate this effort. Silver nanoparticles (AgNPs) kill a broad range of multidrug-resistant DFU isolates by disrupting membranes, generating ROS, blocking respiratory enzymes, and damaging DNA; when held within BC, they release silver ions gradually, which helps limit the cytotoxicity seen at high silver concentrations (Barud et al., 2011; Gupta et al., 2020; Ibraheem & Shaker, 2025). Zinc oxide nanoparticles (ZnO NPs) act against S. aureus and E. coli and also release Zn²⁺ ions, which are cofactors for MMPs and DNA polymerases and help to dampen inflammation, strengthen antioxidant enzymes such as superoxide dismutase and catalase, and promote fibroblast migration and neovascularisation (Khalid et al., 2017; Luo et al., 2020; Melnikova et al., 2021). Copper-based frameworks and graphene oxide (GO) nanosheets add further options, and combinations of BC with phytomedicines such as curcumin or Centella asiatica extract, or with metal–organic frameworks such as ZIF-8 and Cu@ZIF, have been reported to suppress pro-inflammatory cytokines, raise VEGF, PDGF, and CD31, and accelerate closure in diabetic models (Gupta et al., 2020; Khalid et al., 2026; Sajjad et al., 2025; Wang et al., 2024).

Despite this preclinical promise, several questions remain open: how precisely nanoparticle synthesis can be controlled, how stable the particle–fibre interface is, how release behaves in a protease-rich wound, how safe these materials are over the long term, and how readily any of this translates to human patients (Gürsoy et al., 2025; Kushwaha et al., 2022). A structured synthesis of how BC nanocomposites act across the pathophysiological continuum of the DFU is therefore timely.

This review critically evaluates recent developments, therapeutic mechanisms, and translational prospects of BC nanocomposite dressings loaded with antimicrobial nanoparticles for DFU management. Its specific objectives are: To describe the pathophysiological barriers to DFU healing, including persistent hyperglycaemia, oxidative stress, M1 macrophage-dominated inflammation, impaired angiogenesis, and polymicrobial biofilm formation. To evaluate the structural, physical, and mechanical attributes of BC, including its 3D nanofibrillar architecture, water retention, biocompatibility, exudate handling, and role as an ECM-mimetic substrate. To compare fabrication and modification strategies (in situ biosynthesis, ex situ absorption, chemical crosslinking, and green synthesis) for incorporating antimicrobial nanoparticles into BC. To examine the antimicrobial and antibiofilm mechanisms of metallic (AgNPs), metal oxide (ZnO NPs), copper-based, and carbonaceous (GO) nanostructures against common multidrug-resistant DFU pathogens. To synthesize in vitro and in vivo preclinical outcomes, with emphasis on inflammatory cytokines, angiogenic markers, re-epithelialization, and wound closure. To outline translational challenges and future directions, including nanotoxicity, release kinetics, scalable manufacturing, regulation, and stimuli-responsive smart dressings.

2. Nanoparticle-Functionalised Bacterial Cellulose Dressings for Antimicrobial Repair of Diabetic Foot Ulcers

2.1. Pathophysiological Barriers to Diabetic Foot Ulcer Healing

2.1.1. Clinical Burden and Systemic Drivers

DFUs are among the most debilitating, costly, and recurrent complications of diabetes. The lifetime incidence of foot ulceration ranges from 19% to 34%, about 65% of ulcers recur within three to five years, and the lifetime amputation rate approaches 20% (Armstrong et al., 2017). The persistence of these wounds reflects a cluster of systemic and local abnormalities, notably peripheral neuropathy, peripheral arterial disease, persistent microvascular ischaemia, and metabolic dysregulation sustained by hyperglycaemia (Kushwaha et al., 2022; Renuka et al., 2023). Figure 1 summarises how these factors converge on the ulcer bed.

2.1.2. Arrested Inflammation and Oxidative Stress

Normal cutaneous repair proceeds through four overlapping phases: haemostasis, acute inflammation, proliferation with granulation tissue formation and neovascularisation, and ECM remodelling. In the diabetic wound this sequence becomes uncoupled. Prolonged hyperglycaemia promotes AGE accumulation and intracellular ROS generation, and the resulting toxic microenvironment locks the wound in a chronic pro-inflammatory state (Figure 1) (Renuka et al., 2023). Central to this arrest is the failure of macrophages to shift from the M1 to the M2 phenotype. The wound bed

Table 1. Physicochemical and functional properties of bacterial cellulose (BC) matrices relevant to diabetic foot ulcer dressings. The table compiles the principal structural, hydration, mechanical, and biological properties reported for pristine and modified BC across the included studies. For each parameter, the reported range or representative value is given alongside its functional significance for chronic and diabetic wound management and the analytical method used to obtain it. Values are reproduced as reported and were not pooled; differences reflect variations in strain, culture conditions, hydration state, and post-processing. Abbreviations: AFM, atomic force microscopy; BET, Brunauer–Emmett–Teller; BJH, Barrett–Joyner–Halenda; ECM, extracellular matrix; FE-SEM, field-emission scanning electron microscopy; HDF, human dermal fibroblasts; PXRD, powder X-ray diffraction; WVTR, water vapour transmission rate.

Table 2. Nanoparticle functionalisation strategies used to construct antimicrobial bacterial cellulose (BC) and related cellulose nanocomposites. The table compares the fabrication routes used to incorporate metallic, metal oxide, framework, magnetic, and carbonaceous nanostructures into BC or closely related cellulose matrices. For each formulation it lists the synthetic route, the chemical or physical mechanism holding the particles in place, the resulting particle size and morphology, and the main functional advantage for wound care. Route choice governs particle distribution, leaching risk, and release kinetics, and therefore influences the antimicrobial and regenerative outcomes summarised in Tables 3 and 4. Abbreviations: BDP, betulin diphosphate; CNC, cellulose nanocrystal; FCP, fish collagen peptides; HPβCD, hydroxypropyl-β-cyclodextrin; MA, maleic anhydride; ZIF, zeolitic imidazolate framework.

accumulates IL-1α, interleukin-1β (IL-1β), IL-6, and TNF-α, and this sustained cytokine signalling drives over-secretion of MMPs, which degrade newly synthesised collagen and fibronectin and cleave endogenous growth factors (Khalid et al., 2026; Sajjad et al., 2025).

2.1.3. Impaired Angiogenesis and Matrix Degradation

Diabetic wounds also show marked microvascular rarefaction and endothelial dysfunction. VEGF, basic fibroblast growth factor (bFGF), and PDGF are deficient or rapidly degraded, so the regenerating tissue receives too little oxygen and too few nutrients to support capillary sprouting (Singla et al., 2017, 2018). The combined effect of protease excess and growth factor deficiency is a thin, poorly organised granulation bed that closes slowly, if at all.

2.1.4. Microbial Bioburden and Biofilm Formation

The moist, nutrient-rich, and immunologically compromised ulcer bed is an ideal niche for opportunistic pathogens. More than half of chronic DFUs show overt infection, typically involving S. aureus (including methicillin-resistant S. aureus, MRSA), Streptococcus species, P. aeruginosa, and E. coli (Kushwaha et al., 2022; Zheng et al., 2020). These organisms secrete an EPS matrix that binds colonies into resilient biofilms. Biofilm-embedded bacteria can tolerate antibiotic concentrations up to 1,000-fold higher than their planktonic counterparts and evade phagocytosis by neutrophils and macrophages (Kart et al., 2024). Persistent biofilms fuel necrosis, precipitate osteomyelitis, and raise the risk of amputation.

2.1.5. Limitations of Conventional Dressings

Dry gauze, paraffin tulles, and basic polymeric foams do not address these barriers. They tend to adhere to granulation tissue and strip it on removal, manage exudate poorly, offer limited protection against microbial ingress, and lack bioactive or bactericidal function (Revin et al., 2023; Zheng et al., 2020). These shortcomings explain the interest in multifunctional nanostructured dressings that can suppress infection, resolve inflammation, and support tissue regeneration at the same time (Gürsoy et al., 2025; Renuka et al., 2023).

2.2. Bacterial Cellulose as an ECM-Mimetic Biomaterial Matrix

2.2.1. Biosynthesis and Chemical Purity

Among the biopolymers studied for chronic wound care, BC, also called bacterial nanocellulose, stands out for its structure and physical behaviour (Czaja et al., 2007; Revin et al., 2023). It is produced by oxidative fermentation by non-pathogenic, strictly aerobic Gram-negative bacteria such as Komagataeibacter xylinus (formerly Gluconacetobacter xylinus) and related genera, and accumulates as a gelatinous pellicle at the air–liquid interface of the culture. Chemically, BC consists of linear β-(1→4)-linked D-glucose chains, the same primary structure as plant cellulose, but it is free of lignin, pectin, and hemicellulose. Because no harsh fractionation is required, BC shows excellent biocompatibility, low immunogenicity, and good cytocompatibility with human dermal fibroblasts and keratinocytes (Galateanu et al., 2015; Revin et al., 2023).

2.2.2. Three-Dimensional Nanofibrillar Architecture

The defining feature of BC is its ultrafine 3D reticulated network. Ribbon-like microfibrils 10–100 nm in diameter, themselves assembled from elementary fibrils of about 1.5–5 nm, interweave randomly into a dense, highly porous mesh stabilised by extensive hydrogen bonding (Table 1; Figure 2A) (Czaja et al., 2007; Gürsoy et al., 2025). This architecture closely reproduces the nanoscale topology and fibrous organisation of the dermal ECM, which favours cell attachment, migration, and proliferation (Khalid et al., 2026; Yang, 2021).

2.2.3. Fluid Handling, Mechanical Integrity, and Atraumatic Contact

The large surface area and capillary porosity of BC translate into practical advantages at the wound surface (Table 1). BC retains up to 99 wt.% water and can absorb many times its dry mass in fluid, allowing it to take up heavy exudate while keeping the wound moist, which in turn supports cell migration and re-epithelialisation (Melnikova et al., 2021; Zheng et al., 2020). In the hydrated state it also provides balanced water vapour transmission and gas permeation, protecting against both desiccation and maceration of the surrounding skin (Ciecholewska-Juśko et al., 2021; Luo et al., 2020). Its high crystallinity, typically 60–80%, and high degree of polymerisation give hydrated membranes considerable wet tensile strength and flexibility, so they conform to the contours of the foot, resist shear, and form a physical barrier to secondary contamination (Revin et al., 2023; Yang, 2021). Finally, its hydrophilic, non-adherent surface

Figure 1. Pathophysiological barriers that converge to produce a chronic, non-healing diabetic foot ulcer. The upper strip shows the normal sequence of cutaneous repair and marks the inflammatory phase at which diabetic wounds typically stall. Six groups of interacting barriers surround the ulcer: metabolic and oxidative stress (hyperglycaemia, advanced glycation end-products, reactive oxygen species), immune dysregulation (failed M1-to-M2 macrophage switching and persistent IL-1α, IL-1β, IL-6, and TNF-α), matrix and vascular failure (MMP excess and deficient VEGF, bFGF, and PDGF), polymicrobial biofilm, neuro-vascular deficits, and the limitations of passive dressings. Together these drive deep infection, osteomyelitis, sepsis, and amputation. Constructed from Armstrong et al. (2017), Kushwaha et al. (2022), Renuka et al. (2023), and Section 2.1.

Figure 2. Bacterial cellulose as an ECM-mimetic matrix and the principal routes used to convert it into an antimicrobial nanocomposite. (A) Pristine BC secreted by Komagataeibacter xylinus forms a lignin- and pectin-free 3D network of 10–100 nm fibrils; the surrounding tiles summarise the reported water-holding capacity, vapour transmission, crystallinity, porosity, mechanical strength, and cytocompatibility (values from Table 1). Its key limitation, the absence of intrinsic antimicrobial activity, is highlighted. (B) The four fabrication routes (ex situ immersion, in situ biosynthesis, chemical modification, and green synthesis) are shown with representative examples and their main advantages (+) and drawbacks (−).

allows BC to be applied and removed without damaging the advancing epithelial margin, sparing patients additional pain during dressing changes (Czaja et al., 2007).

2.2.4. The Antimicrobial Gap of Pristine BC

These advantages come with a significant caveat. Pristine BC has no intrinsic antimicrobial, antibiofilm, or antioxidant activity. Placed over a heavily contaminated ulcer, unmodified BC cannot inhibit bacterial growth, and microorganisms may even colonise its porous mesh (Zheng et al., 2020). Functionalising BC with antimicrobial nanoscale agents has therefore become an essential design strategy (Ibraheem & Shaker, 2025; Revin et al., 2023).

2.3. Fabrication and Modification Strategies for BC Nanocomposites

The reactive hydroxyl groups on BC nanofibrils and the open interstitial network make BC an effective template for organic and inorganic nanostructures. Four broad fabrication routes have been used to construct antimicrobial BC nanocomposites (Table 2; Figure 2B).

2.3.1. Ex Situ Incorporation by Physical Absorption and Immersion

The ex situ approach is the most widely used because it is simple, reproducible, and leaves the crystalline BC matrix intact. Purified wet pellicles or freeze-dried aerogels are immersed in colloidal suspensions of pre-synthesised nanoparticles, bioactive molecules, or polymers under agitation or sonication. Particles penetrate the mesh by capillary action and are retained by physical entrapment and non-covalent interactions, including hydrogen bonding, electrostatic attraction, and van der Waals forces (Zheng et al., 2020). Khalid et al. (2026) used this approach to disperse GO nanosheets evenly through the BC network; Fourier-transform infrared spectroscopy (FTIR) and scanning electron microscopy (SEM) indicated close interpenetration mediated by hydrogen bonding between hydroxyl and carboxyl groups. Gupta et al. (2020) similarly loaded purified BC hydrogels with curcumin–cyclodextrin-functionalised AgNPs and achieved high loading without loss of hydration or mechanical stability.

2.3.2. In Situ Biosynthesis

In the in situ route, nanomaterials, polymers, or reinforcing additives are added directly to the fermentation medium. As K. xylinus extrudes cellulose protofibrils, the additives become trapped within the growing network, yielding a homogeneous distribution through the full thickness of the pellicle. Galateanu et al. (2015) produced BC–magnetite (Fe₃O₄) nanocomposites in this way and showed that the resulting scaffolds supported stem cell proliferation. The approach requires careful optimisation, however, because high concentrations of metal precursors or antimicrobial agents can be toxic to the producing bacteria, lowering cellulose yield and altering fibril crystallisation (Revin et al., 2023).

2.3.3. Chemical Surface Modification and Crosslinking

Chemical functionalisation can strengthen interfacial bonding, regulate swelling, and prevent burst release of loaded particles. Options include TEMPO-mediated oxidation to generate carboxylate groups, esterification, etherification, and polymer grafting (Zheng et al., 2020). Luo et al. (2020) modified BC with maleic anhydride (MA) to introduce carboxyl anchoring sites for in situ ZnO deposition; the resulting membrane was flexible and porous, resisted particle agglomeration, and retained a water vapour transmission rate (WVTR) of 2,856.60 g/m²/day. Hybridising cellulose derivatives such as carboxymethyl cellulose (CMC) with poly(vinyl alcohol) (PVA) by citric acid crosslinking has likewise been used to anchor metallic nanoparticles in stable, non-leaching, biocompatible hydrogels (Mansur et al., 2023). Crosslinking can also convert BC into superabsorbent forms suited to highly exudative chronic wounds (Ciecholewska-Juśko et al., 2021).

2.3.4. Green and Bio-Mediated Nanoparticle Synthesis

Concerns about hazardous reducing agents such as sodium borohydride and hydrazine, and about residual solvents, have encouraged green synthesis. Plant extracts rich in polyphenols, flavonoids, and reducing sugars serve as both reducing and capping agents. Singla et al. (2017) used Syzygium cumini leaf extract to synthesise AgNPs directly on cellulose nanocrystals (CNCs), avoiding toxic residues and producing cytocompatible, strongly antibacterial nanobiocomposites; a related bamboo-derived system followed (Singla et al., 2018). Curcumin and Centella asiatica extract have also been incorporated into cellulose- and polymer-based nanocomposites to mediate metal reduction, strengthen antioxidant defences, and stimulate proliferation (Gupta et al., 2020; Wang et al., 2024).

2.4. Mechanistic Antimicrobial and Bioactive Actions of Embedded Nanoparticles

Incorporating nanoscale agents converts BC from a passive cover into an active therapeutic platform able to attack drug-resistant biofilms and alter the pathological wound microenvironment (Figure 3; Table 3).

2.4.1. Silver Nanoparticles

AgNPs are the most extensively studied antimicrobial agents in wound care, owing to their broad-spectrum activity against multidrug-resistant pathogens including S. aureus, MRSA, and P. aeruginosa (Barud et al., 2011; Ibraheem & Shaker, 2025). Within BC, AgNPs act through several overlapping mechanisms. They adhere electrostatically to negatively charged lipopolysaccharide or peptidoglycan, pitting the membrane and causing leakage of intracellular contents. In the aqueous wound environment they continuously release Ag⁺ ions, which bind sulfhydryl groups on respiratory chain enzymes and structural proteins, interrupting electron transport and ATP synthesis. Intracellular silver also catalyses ROS production, leading to DNA condensation, inhibition of replication, and cell lysis. Beyond direct killing, AgNPs can inactivate excess MMPs and so limit inflammatory tissue breakdown (Gupta et al., 2020; Zheng et al., 2020). Immobilisation in the BC network also addresses the longstanding concern about silver toxicity: uncontrolled silver can induce keratinocyte apoptosis and slow re-epithelialisation, whereas the BC matrix acts as a reservoir that keeps local Ag⁺ within the bactericidal window but below the threshold that harms fibroblasts (Gupta et al., 2020; Singla et al., 2017).

2.4.2. Zinc Oxide Nanoparticles and Zeolitic Imidazolate Frameworks

ZnO NPs are an attractive alternative or complement to silver, combining lower cytotoxicity, chemical stability, low cost, and genuine biological functions (Khalid et al., 2017; Luo et al., 2020). Their antibacterial activity arises from electrostatic adhesion to the cell wall, mechanical destabilisation, and ROS generation (hydroxyl radicals and hydrogen peroxide), which drives lipid peroxidation and cell death. ZnO NPs also act as a slow-release source of Zn²⁺, a cofactor for more than 200 enzymes, including alkaline phosphatase, carbonic anhydrases, and DNA and RNA polymerases. Released zinc stimulates fibroblast proliferation and migration, supports keratinocyte re-epithelialisation, and moderates local inflammatory cytokine secretion. Melnikova et al. (2021) showed that BC–ZnO dressings increased superoxide dismutase (SOD) and catalase activity and reduced malondialdehyde (MDA), indicating reversal of oxidative stress. Zeolitic imidazolate framework-8 (ZIF-8) loaded onto BC has been reported to release Zn²⁺ in a controlled, sustained manner, supporting angiogenic sprouting while suppressing E. coli and S. aureus (Gürsoy et al., 2025).

2.4.3. Copper Nanostructures and Metal–Organic Frameworks

Copper nanostructures combine potent antimicrobial activity with a pronounced capacity to stimulate angiogenesis. Sajjad et al. (2025) designed a BC hydrogel membrane containing copper-doped, fish collagen peptide-loaded zeolitic imidazolate frameworks (Col-Cu@ZIF). In infected diabetic wounds, released Cu²⁺ reacted with endogenous hydrogen peroxide in a Fenton-like reaction, generating localised ROS that rapidly disrupted biofilms. The sustained co-delivery of copper ions and collagen peptides also stimulated endothelial migration, increased CD31 and VEGF expression, and promoted dense collagen remodelling.

2.4.4. Graphene Oxide and Carbonaceous Nanostructures

GO nanosheets provide mechanical reinforcement as well as contact-mediated killing. Their sharp edges can cleave bacterial membranes, a so-called "nano-knife" effect, while generating localised oxidative stress (Zheng et al., 2020). In a diabetic mouse model, Khalid et al. (2026) found that BC–GO films produced inhibition zones of 24.6 mm against S. aureus and 23 mm against E. coli and achieved complete wound closure within 21 days, outperforming a commercial ointment.

2.5. In Vitro and In Vivo Biological Outcomes in Diabetic Wound Repair

Preclinical studies suggest that antimicrobial BC dressings accelerate diabetic wound repair through several linked cellular and molecular routes (Table 4; Figure 3; Figure 4).

2.5.1. Suppression of Chronic Inflammation and Oxidative Stress

Real-time reverse-transcription PCR and immunohistochemistry show that functionalised nanocomposites reduce gene and protein expression of IL-1α, IL-1β, IL-6, and TNF-α in diabetic wound tissue. Khalid et al. (2026) reported reduced IL-1α and TNF-α expression

and less inflammatory cell infiltration in BC–GO-treated diabetic mice than in untreated controls. Singla et al. (2017, 2018) found that CNC–AgNP nanobiocomposites lowered IL-6 and TNF-α over 18 days in streptozotocin (STZ)-induced diabetic mice. By damping the inflammatory cascade and scavenging free radicals, these dressings appear to let stalled wounds move on into active proliferation.

2.5.2. Induction of Angiogenesis and Neovascularisation

Restoring a functional capillary bed is essential for reperfusing ischaemic ulcers. Nanoparticle-loaded BC dressings increase VEGF, PDGF, and the endothelial marker CD31 (Sajjad et al., 2025). Sajjad et al. (2025) observed dense CD31-positive capillary networks in diabetic mice treated with Col-Cu@ZIF–BC membranes, and Singla et al. (2017, 2018) documented significant increases in VEGF, bFGF, and PDGF with silver-functionalised nanocellulose, accompanied by greater vessel density and better granulation tissue.

2.5.3. Collagen Organisation, Re-Epithelialisation, and Wound Closure

Macroscopic and histological assessments indicate that functionalised dressings speed closure substantially. In STZ-diabetic mice, wounds treated with pristine controls or commercial ointments typically reached 88–92% closure with persistent purulent exudate by days 18–21, whereas wounds treated with BC–AgNP, BC–GO, or Col-Cu@ZIF–BC dressings reached 98–100% closure over the same period (Figure 4A) (Khalid et al., 2026; Sajjad et al., 2025; Singla et al., 2018). Masson's trichrome staining showed thick, aligned, mature collagen bundles in treated wounds, in contrast to the fragmented, randomly oriented fibrils of untreated diabetic wounds, and sustained antimicrobial release combined with a moist environment encouraged keratinocyte migration and regeneration of hair follicles and sebaceous glands (Luo et al., 2020; Singla et al., 2018).

2.6. Translational Challenges Identified in the Literature

2.6.1. Nanoparticle Toxicity and Long-Term Biocompatibility

The possibility of systemic accumulation and cellular toxicity from inorganic nanoparticles, particularly AgNPs and metal oxides, remains a concern. High intracellular metal ion uptake can cause oxidative DNA damage and mitochondrial apoptosis in healthy dermal cells. Standardised dose–response thresholds, controlled release kinetics, and stable chelation within the BC mesh will be needed to establish long-term safety (Ibraheem & Shaker, 2025; Kushwaha et al., 2022).

2.6.2. Standardisation and Large-Scale Biomanufacturing

Fermentation of BC by K. xylinus varies from batch to batch in membrane thickness, fibre density, pore size, and strength. Industrial-scale production will require tightly controlled bioreactors, inexpensive carbon sources such as agro-industrial by-products, and validated, automated protocols for nanoparticle synthesis and loading (Gürsoy et al., 2025; Revin et al., 2023).

2.6.3. Wound Fluid Complexity and Smart-Dressing Integration

Human DFU exudate is more proteolytic and variable in pH than that of rodent excision wounds, which may alter nanoparticle dissolution, cause premature leaching, or allow a protein corona to mask active sites (Kushwaha et al., 2022). Looking ahead, stimuli-responsive designs that release antimicrobials in response to pH shifts, temperature, or bacterial enzymes, together with chromogenic or bioelectronic sensors embedded in BC, could permit on-demand therapy and real-time monitoring without removing the dressing (Stanescu et al., 2021; Yang, 2021).

3. Methods

3.1. Information Sources and Search Strategy

Four electronic databases were searched: PubMed/MEDLINE, Scopus, Web of Science Core Collection, and Google Scholar (first 200 results, sorted by relevance). The search covered publications from January 2007 to March 2026, a window chosen to capture the earliest biomedical assessments of microbial cellulose (Czaja et al., 2007) through the most recent diabetic wound studies (Khalid et al., 2026). The core PubMed query combined Medical Subject Headings (MeSH) and free-text terms as follows:

("bacterial cellulose" OR "bacterial nanocellulose" OR "microbial cellulose" OR "Komagataeibacter" OR "cellulose nanocrystals") AND ("silver nanoparticles" OR "zinc oxide" OR "copper" OR "graphene oxide" OR "metal-organic framework" OR "nanoparticle*" OR "nanocomposite*") AND ("Diabetic Foot"[MeSH] OR "diabetic wound*" OR "chronic wound*" OR "Wound Healing"[MeSH] OR "Bandages"[MeSH] OR "wound dressing*")

The string was adapted to the syntax of each database (for example, TITLE-ABS-KEY fields in Scopus and TS fields in Web of Science). Reference lists of all included articles and of key reviews (Gürsoy et al., 2025; Ibraheem & Shaker, 2025; Revin et al., 2023; Zheng et al., 2020) were hand-searched to identify additional eligible studies. Searches were restricted to English-language publications.

3.2. Eligibility Criteria

Eligibility was defined using a Population–Intervention–Comparator–Outcome (PICO) framework. Studies were included if they (a) fabricated BC, or a closely related cellulose-based or polymeric nanocomposite tested specifically in diabetic wounds, functionalised with antimicrobial nanoparticles or nanostructures; (b) reported physicochemical characterisation (for example, SEM, FTIR, X-ray diffraction, WVTR, or water-holding capacity), antimicrobial or antibiofilm testing against wound-relevant pathogens, cytocompatibility testing, or in vivo wound healing outcomes; and (c) were original research articles, doctoral theses with full experimental data, or peer-reviewed reviews providing epidemiological, mechanistic, or translational context. Studies were excluded if they (a) did not involve a wound-dressing application; (b) used non-antimicrobial cellulose formulations only, unless they served as an unmodified BC comparator; (c) were conference abstracts, editorials, or patents without full data; or (d) duplicated data reported elsewhere. Non-BC nanocellulose and polymer platforms tested in diabetic models (Alzahrani et al., 2024; Mansur et al., 2023; Singla et al., 2017, 2018; Wang et al., 2024) were retained deliberately as performance benchmarks.

3.3. Study Selection

Records were exported to a reference manager and de-duplicated. Two reviewers independently screened titles and abstracts against the eligibility criteria, and then assessed the full texts of potentially relevant records. Disagreements were resolved by discussion and, where necessary, by a third reviewer. A total of 25 sources met the criteria and were included: 16 primary experimental studies, 8 reviews or clinical overviews, and 1 doctoral thesis. The selection process is reported in a PRISMA 2020 flow diagram (Supplementary Figure S1).

3.4. Data Extraction

A standardised extraction form was piloted on three studies and then applied to all included records by one reviewer and verified by a second. Extracted variables were organised into four domains corresponding to Tables 1–4: (i) physicochemical properties of the BC matrix, including fibril diameter, water-holding capacity, WVTR, crystallinity, porosity, specific surface area, tensile properties, and cytocompatibility, together with the analytical method used (Table 1); (ii) functionalisation route, nanoparticle type and size, interfacial binding mechanism, and morphology (Table 2); (iii) antimicrobial and antibiofilm testing, including target organism and strain, assay method (disc diffusion, colony-forming unit counts, crystal violet or XTT biofilm assays, broth microdilution), inoculum, and quantitative outcome (Table 3); and (iv) in vivo design, including species and strain, method of diabetes induction and glycaemic threshold, wound type and size, sample size per group, treatment regimen and duration, closure rate, and histological and molecular endpoints (Table 4). Numerical values were recorded exactly as reported, including means and standard deviations; where only graphical data were available, values were not estimated.

4. Structural Performance, Antimicrobial Efficacy, and Diabetic Tissue Regeneration of Bacterial Cellulose Nanocomposites

4.1. Nanofibrillar Architecture and Biophysical Performance of the BC Matrix

4.1.1. Network Morphology and ECM Mimicry

Across the included studies, BC emerges with a physical profile that seems well matched, perhaps unusually so, to the hostile environment of a chronic DFU (Czaja et al., 2007; Revin et al., 2023). Pristine BC from K. xylinus fermentation forms a reticulated 3D network whose ribbon-like fibres consistently measure between 10 and 100 nm (Table 1) (Gürsoy et al., 2025; Yang, 2021). Porosity typically exceeds 80–90% and the specific surface area is above 150 m²/g, values that broadly mirror the porosity and topography of native dermal ECM (Figure 2A) (Khalid et al., 2026; Sajjad et al., 2025; Zheng et al., 2020).

4.1.2. Fluid Management and Vapour Permeability

Exudate handling is where BC most clearly distinguishes itself. Reported water-holding capacity ranges from 98.6% to 99.4% by mass, with uptake of 100–200 times the dry weight (Table 1) (Melnikova et al., 2021; Singla et al., 2017). WVTR values for engineered dressings cluster between 2,500 and 2,856.60 g/m²/day, the upper figure coming from MA-functionalised nano-ZnO/BC membranes (Luo et al., 2020). That range is, in practical terms, high enough to avoid fluid pooling and maceration but not so high that the wound dries out, a balance that occlusive polymer films often miss (Ciecholewska-Juśko et al., 2021; Zheng et al., 2020).

4.1.3. Mechanical Integrity and Cytocompatibility

Mechanically, the picture is a little more varied, as one might expect from materials tested in different hydration states. Crystallinity indices of 59.77% to over 70%, dominated by the metastable Iα allomorph, correspond to wet tensile strengths of about 2.02 ± 0.30 MPa for soft hydrogels and up to 200–260 MPa for dried composite films, with elongation at break between 22% and 63.8% (Table 1) (Gürsoy et al., 2025; Revin et al., 2023; Yang, 2021). Even at the low end, this seems sufficient for the dressing to tolerate plantar shear without tearing or bonding to granulation tissue (Czaja et al., 2007; Luo et al., 2020). Cytocompatibility data were reassuringly consistent: viability in human dermal fibroblasts, keratinocytes, and L929 fibroblasts generally exceeded 85–95%, supporting the view that alkali-purified BC is essentially non-immunogenic (Galateanu et al., 2015; Kart et al., 2024; Melnikova et al., 2021).

4.2. Nanoparticle Functionalisation, Interfacial Stability, and Release Behaviour

4.2.1. Ex Situ Impregnation

Because pristine BC does not kill bacteria, everything depends on how well antimicrobial nanostructures are anchored within it, and the different routes involve real trade-offs (Table 2; Figure 2B). Ex situ immersion and sonication preserved the native crystallinity and hydration of BC while giving homogeneous loading. Curcumin–hydroxypropyl-β-cyclodextrin–capped AgNPs (10–30 nm), for instance, distributed evenly through BC hydrogels, sidestepped curcumin's poor water solubility, and allowed sustained, non-cytotoxic Ag⁺ release (Gupta et al., 2020; Ibraheem & Shaker, 2025). Vacuum-assisted GO immersion produced an interpenetrating, hydrogen-bonded composite of noticeably greater stiffness (Khalid et al., 2026). The weaker, non-covalent anchoring of this route does, however, raise the question of how much material might leach under prolonged exudation, a point that few studies measured directly.

4.2.2. Chemical Anchoring and Framework Entrapment

Chemical functionalisation appears to be the most reliable way to prevent aggregation and burst release. MA modification introduced esterified carboxyl sites that coordinated Zn²⁺ and seeded the in situ nucleation of 30–60 nm ZnO particles in a characteristic "string-beaded" arrangement along the fibrils (Table 2) (Luo et al., 2020). One-pot co-precipitation of Col-Cu@ZIF (80–150 nm) within BC hydrogels gave a dressing that degraded steadily under physiological conditions, releasing Cu²⁺ and collagen peptides into infected exudate (Sajjad et al., 2025). Borate-ester crosslinking of ZnO, ZnO–borax, and ZnCuO₂ particles provided a further variant, with nanoclusters of 20–80 nm held within the interstitial voids (Kart et al., 2024).

4.2.3. Green Synthesis and In Situ Biosynthesis

Phytofabrication with S. cumini or bamboo leaf extracts avoided hazardous reductants altogether and yielded 5–20 nm AgNPs grafted firmly onto CNCs at a silver loading of only 0.05 wt.% (Table 2) (Singla et al., 2017, 2018). At the other end of the spectrum, adding Fe₃O₄ nanoparticles to the fermentation broth produced superparamagnetic composites with particles distributed throughout the matrix and good adhesion of human adipose-derived stem cells (Galateanu et al., 2015). Whether either route can be scaled with consistent particle size is, at present, uncertain.

4.3. Antimicrobial Bioburden Suppression and Biofilm Eradication

4.3.1. Bactericidal Activity Against Planktonic Pathogens

Taken together, the antimicrobial data indicate broad-spectrum bactericidal and antibiofilm activity against the pathogens most often isolated from infected DFUs (Table 3; Figure 3). In suspension and colony-count assays, the 5 wt.% nano-ZnO/BC membrane reduced viable S. aureus and E. coli by more than 99.9% within 24 h, apparently through membrane destabilisation and local ROS generation (Luo et al., 2020). Col-Cu@ZIF–BC hydrogels gave a >99.5% reduction in counts relative to untreated controls; here, Cu²⁺-driven Fenton-like chemistry converted endogenous H₂O₂ into hydroxyl and superoxide radicals that damaged membranes and broke down EPS (Sajjad et al., 2025).

4.3.2. Contact-Mediated Killing by Graphene Oxide

In disc diffusion assays, BC–GO films produced inhibition zones of 24.6 ± 0.5 mm against S. aureus and 23.0 ± 1.0 mm against E. coli, significantly larger than those of pristine BC or a commercial Polyfax control (Table 3) (Khalid et al., 2026). The effect is attributed to physical shearing by GO edges combined with oxidative stress (Khalid et al., 2026; Zheng et al., 2020), although inhibition zones from a contact-active material are not easy to compare with those from ion-releasing silver or zinc systems.

4.3.3. Activity Against Preformed Biofilms and Fungal Pathogens

Biofilm data are fewer, but encouraging. Crystal violet and XTT assays showed that BC–borax, BC–ZnO–borax, and BC–ZnO dressings degraded mature S. aureus biofilms by 71.74%, 66.60%, and 65.53%, respectively, and BC–ZnCuO₂ disrupted preformed P. aeruginosa biofilms (Figure 4C) (Kart et al., 2024). Curcumin–AgNP-loaded BC hydrogels extended protection to the multidrug-resistant yeast Candida auris without apparent fibroblast toxicity (Gupta et al., 2020). Comparable silver-loaded CMC/PVA hydrogels produced concentration-dependent inhibition of reference S. aureus and E. coli strains (Mansur et al., 2023), and early BC–silver membranes had already shown clear antibacterial activity more than a decade earlier (Barud et al., 2011). It should be said that roughly 65–72% biofilm inhibition, while meaningful, is still some distance from eradication.

4.4. Accelerated Diabetic Wound Closure and Tissue Regeneration In Vivo

4.4.1. Modulation of Chronic Inflammation and Oxidative Stress

In STZ-induced diabetic mice, wounds dressed with gauze or vehicle remained in a prolonged inflammatory state, with dense neutrophil infiltration and high pro-inflammatory cytokine levels (Khalid et al., 2026; Singla et al., 2018). BC–GO films reduced IL-1α and TNF-α mRNA at days 7 and 14, suggesting an earlier exit from the inflammatory phase (Table 4) (Khalid et al., 2026). The CNC–AgNP systems NC-1 and DH/BB-CNC-Ag similarly lowered serum and tissue IL-6 and TNF-α over 18 days (Singla et al., 2017, 2018). BC–ZnO–betulin diphosphate dressings raised SOD and catalase activity and lowered MDA, consistent with reduced free-radical damage (Melnikova et al., 2021), while Col-Cu@ZIF–BC membranes promoted the M1-to-M2 macrophage transition (Sajjad et al., 2025).

4.4.2. Neovascularisation and Endothelial Sprouting

The compromised microcirculation of the diabetic wound seems to respond well to sustained ion and peptide release. Immunohistochemistry of Col-Cu@ZIF–BC-treated wounds showed intense CD31 staining and extensive capillary lumen formation in the granulation bed (Sajjad et al., 2025). Molecular profiling after treatment with silver-functionalised CNCs revealed significant increases in VEGF, bFGF, and PDGF (Table 4) (Singla et al., 2017, 2018), which presumably helped restore perfusion to the regenerating dermis (Singla et al., 2018).

4.4.3. Collagen Maturation, Re-Epithelialisation, and Closure Rates

Planimetry consistently showed faster closure with functionalised dressings than with pristine BC, gauze, or commercial ointments (Table 4; Figure 4A–B). BC–GO films reached 100% closure by day 21, with a residual area at day 7 of 18.6 ± 2.19 mm², compared with 57 ± 6.63 mm² for pristine BC and 97.66 mm² for Polyfax (Figure 4B) (Khalid et al., 2026). Col-Cu@ZIF–BC hydrogels achieved more than 95% contraction by days 12–14, outperforming a recombinant human epidermal growth factor gel (Sajjad et al., 2025). The CNC–AgNP nanobiocomposites gave 98–100% closure by day 18, against 88–92% in untreated controls, which still showed purulent exudate (Singla et al., 2017, 2018). For comparison, coaxial fusidic acid/pirfenidone nanofibres achieved near-complete closure by days 11–13, ahead of a povidone-iodine dressing (Alzahrani et al., 2024), and poloxamer/ZnO composites containing C. asiatica extract improved contraction and collagen deposition in diabetic rats (Wang et al., 2024). Histologically, treated wounds showed thick, stratified, well-aligned collagen with high hydroxyproline content, in contrast to the loose, fragmented fibres of controls (Khalid et al., 2026; Singla et al., 2017, 2018), and faster migration of the epithelial tongue, with regenerated hair follicles and skin appendages (Luo et al., 2020; Singla et al., 2018).

5. Bridging Preclinical Promise and Clinical Reality for Bacterial Cellulose Nanocomposite Dressings

5.1. Antimicrobial Nanoparticle-Loaded Bacterial Cellulose Dressings Target Multiple Barriers to Diabetic Foot Ulcer Healing

The evidence assembled here points, fairly consistently, in one direction: adding antimicrobial nanoparticles to BC turns a well-behaved but biologically inert dressing into one that appears to act on several DFU barriers at once (Figure 3). The matrix itself contributes moisture balance, vapour permeability, conformability, and an ECM-like scaffold (Table 1). The nanoparticles contribute bacterial killing and, perhaps more interestingly, signals that appear to nudge the wound out of its inflammatory stall (Tables 3 and 4). It is tempting to read this as a solved problem. It is not, or at least not yet, and much of what follows concerns the distance between these rodent results and an infected plantar ulcer in an older patient with peripheral arterial disease.

5.2. Antimicrobial Potency Versus Host Cell Safety

Silver remains the benchmark, and its broad-spectrum activity is not seriously in doubt (Barud et al., 2011; Ibraheem & Shaker, 2025). What is less settled is the therapeutic window. The BC network evidently slows Ag⁺ release enough to protect fibroblasts in vitro (Gupta et al., 2020), and the green-synthesised CNC systems achieved strong effects at only 0.05 wt.% silver (Singla et al., 2017). But the included studies rarely reported cumulative release profiles over the full treatment period, and almost none measured tissue or systemic silver. Zinc- and copper-based systems may offer a different balance; their ions are physiological cofactors, and the antioxidant and pro-angiogenic responses reported for BC–ZnO and Col-Cu@ZIF–BC hint at benefits that go beyond killing bacteria (Melnikova et al., 2021; Sajjad et al., 2025). Even so, essential trace elements can be toxic at local concentrations, and the dose–response relationships for chronic exposure in human skin remain largely unknown (Kushwaha et al., 2022). GO adds a contact-mediated mechanism that should be less prone to classical resistance (Khalid et al., 2026), though its long-term fate in tissue has not been studied in this setting.

5.3. Fabrication Route as a Determinant of Performance

One finding that tends to be underplayed is how strongly the fabrication route shapes behaviour (Table 2; Figure 2B). Ex situ loading is simple and preserves BC structure, but physically entrapped particles may migrate, and leaching was rarely quantified (Gupta et al., 2020; Khalid et al., 2026). Chemical anchoring, as with MA-modified BC, gave more uniform particles and retained good vapour transmission (Luo et al., 2020), at the cost of extra reagents and processing steps that a regulator will scrutinise. In situ biosynthesis gives the most homogeneous distribution but may suppress cellulose production when antimicrobial precursors are present (Galateanu et al., 2015; Revin et al., 2023). Green synthesis is attractive on safety and sustainability grounds, yet natural extracts vary in composition, which could translate into variable particle size and dose (Singla et al., 2017, 2018). It would be useful if future studies compared routes directly, with the same nanoparticle and the same wound model, rather than leaving readers to infer differences across laboratories.

5.4. The Gap Between Rodent Models and Human Diabetic Foot Ulcers

Perhaps the most important caveat concerns the models themselves. Most in vivo data come from STZ-induced diabetic mice or rats with small, acute, full-thickness excisional wounds of 6–15 mm, followed for 14–21 days, with group sizes of 4 to 12 animals (Table 4). Rodent skin heals substantially by contraction rather than re-epithelialisation, whereas human DFUs are chronic, often ischaemic, frequently neuropathic, and colonised by mature polymicrobial biofilms that have had weeks or months to form (Armstrong et al., 2017; Kushwaha et al., 2022). Only a few studies used an infected model (Luo et al., 2020; Sajjad et al., 2025), and none, as far as we could determine, examined repeated application over clinically relevant periods. Human exudate is also more proteolytic and variable in pH, which might accelerate dissolution, cause premature leaching, or allow a protein corona to blunt antimicrobial surfaces. The closure rates of 95–100% (Figure 4A) should therefore be read as evidence of biological activity rather than as predictions of clinical healing rates.

5.5. Comparison with Non-BC Nanocomposite Platforms

The benchmark studies included in this review are informative. CNC–AgNP composites (Singla et al., 2017, 2018), silver-loaded CMC/PVA hydrogels (Mansur et al., 2023), poloxamer/ZnO with C. asiatica (Wang et al., 2024), and antibiotic-loaded coaxial nanofibres (Alzahrani et al., 2024) all achieved broadly similar outcomes in diabetic models. This suggests that the antimicrobial or bioactive payload, rather than the BC matrix alone, drives much of the effect. BC's distinctive contribution may lie in

Figure 3. Antimicrobial and pro-regenerative mechanisms of nanoparticles embedded in bacterial cellulose dressings. Each column represents one nanoparticle class: silver nanoparticles, zinc oxide nanoparticles and ZIF-8, copper-doped zeolitic imidazolate frameworks with collagen peptides, and graphene oxide. The upper panels list antimicrobial and antibiofilm actions (membrane disruption, ion release, ROS and Fenton-like chemistry, contact killing). The lower panels list host-directed actions (MMP inactivation, antioxidant enzyme induction, macrophage repolarisation, angiogenesis, collagen remodelling). These converge on accelerated diabetic wound repair. Based on Sections 2.4 and 2.5 and Tables 3 and 4.

Figure 4. Quantitative preclinical outcomes reported for antimicrobial nanocellulose dressings. (A) Reported wound closure endpoints in diabetic rodent models, with the assessment day in parentheses; bars show the reported value or lower bound and whiskers the reported range, and the axis begins at 80% (Khalid et al., 2026; Sajjad et al., 2025; Singla et al., 2017, 2018). (B) Residual wound area at day 7 for BC–GO film versus pristine BC and Polyfax, mean ± SD where reported (Khalid et al., 2026). (C) Inhibition of preformed S. aureus biofilms by borax- and ZnO-loaded BC (Kart et al., 2024). Values were plotted as reported, without pooling; studies differ in model and design.

its handling and fluid-management properties (Table 1), which matter clinically but are seldom measured as endpoints in animal studies. Chitosan–BC biocomposites carrying polymeric nanoparticles (Stanescu et al., 2021) and superabsorbent crosslinked BC (Ciecholewska-Juśko et al., 2021) indicate how the matrix itself might be further tuned for highly exudative ulcers.

5.6. Manufacturing, Regulation, and Smart-Dressing Directions

Scale-up remains a practical obstacle. BC fermentation varies between batches, and the addition of nanoparticles introduces further variables in particle size, loading, and distribution (Gürsoy et al., 2025; Revin et al., 2023). Combination products of this kind, part device and part active agent, are likely to face demanding regulatory review, particularly where novel nanomaterials are involved. On a more optimistic note, BC is well suited to the next generation of responsive dressings. pH-triggered release (wound pH tends to rise during active infection), temperature-sensitive or enzyme-cleavable linkers, and embedded chromogenic or bioelectronic sensors could allow both on-demand antimicrobial delivery and monitoring without dressing removal (Yang, 2021). Whether such complexity can be delivered at a cost that suits high-volume DFU care is, admittedly, an open question.

5.7. Limitations of This study

This review has its own limitations. The evidence base is small and heterogeneous, which precluded meta-analysis. Several included studies used cellulose nanocrystals or non-BC polymers, so not every finding can be attributed to BC specifically. Outcome reporting was inconsistent; many studies gave closure percentages without variance, and few reported blinding or randomisation. Publication bias towards positive results is likely. Finally, the restriction to English-language sources may have excluded relevant work.

6. Conclusion

Bacterial cellulose brings a rare combination of purity, strength, moisture control, and ECM-like architecture to diabetic wound care, but on its own it cannot control the polymicrobial infection that drives so many ulcers towards amputation. The studies reviewed here show that loading BC with silver, zinc oxide, copper-framework, or graphene oxide nanostructures largely fills that gap. These composites killed common DFU pathogens in vitro, disrupted biofilms, reduced inflammatory cytokines, promoted angiogenesis and organised collagen, and accelerated closure in diabetic rodent models within weeks. Still, the evidence is preclinical, heterogeneous, and short in duration. Safe dosing windows, long-term nanoparticle fate, leaching under human wound conditions, and reproducible large-scale manufacture are not yet established. Head-to-head comparisons of fabrication routes, infected chronic-wound models, and carefully designed clinical trials should come next. If those hurdles can be cleared, antimicrobial BC nanocomposites could become a practical, multifunctional option for one of diabetes' most damaging complications.

References