Depending on different context, plasma can be referred to as either the largest part of human blood (physiological plasma) or the fourth state of matter (physical plasma). In physics, plasma consists of electrons, ions, photons, metastables, and electromagnetic fields. Plasma was historically generated at high temperature and under high pressure. Nowadays, cold atmospheric plasma (CAP) could be produced at atmospheric pressure and operate under room temperature.

Various technologies have been used to form atmospheric plasma (1, 2), including, but are not limited to, pulsed atmospheric arc (PAA) technology and piezoelectric direct discharge (PDD) technology (3). For example, a PDD CAP generator ( Figure 1 ) uniquely combines voltage transformation and plasma generation in a single, compact, and highly efficient component (4). Its piezoelectric radiofrequency (RF) plasma generator provides several advantages, such as a high ionization speed, multi-gas ignition, low power consumption, and an effective generation rate of reactive oxygen and nitrogen species (RONS).

The RONS produced by CAP’s hierarchical reaction with ambient air play a key role in various biological and cellular pathways, which has facilitated the application of CAP toward the medical field. In addition, the UV light and transient electric fields might also be important for biomedical applications (5–7). Plasma medicine, a newly coined term, implies an interdisciplinary subject incorporating physics, chemistry, life science, and medicine. The applications of plasma medicine can be broadly divided into direct ones and indirect ones. The former refers to exposing a living recipient under a gaseous plasma plume, and the latter implies delivering the reactive species of the CAP to a living recipient through an intermediate carrier.

The direct medical applications of CAP have been witnessed in several medical specialties. These include, but are not limited to, dentistry (e.g., tooth bleaching and root canal disinfection) (8), neurology (e.g., enhancing neural cell differentiation into neurons both in vitro and in vivo) (9), oncology (e.g., inducing immunogenic cancer cell death and synergy with other anti-neoplastic agents) (10), and infection control (e.g., controlling microbial biofilm) (11), etc. On the other hand, the indirect applications of CAP are achieved through either a solid object (e.g., a surgical implant) (12), a liquid object (e.g., the plasma-activated medium) (13), or a mixture (e.g., stem cell niche) (3). Irrespective of the operating mode, the biomedical effect of CAP penetrates through cellular, protein, and even DNA/mRNA levels (10, 13–15).

Plasma dermatology has emerged as an attractive specialty of plasma medicine, partially because skin is the largest and most superficial organ in the human body, which makes plasma treatment relatively easy ( Figure 1 ). The original series of clinical randomized controlled trials (RCTs) of using CAP in dermatology focused primarily on chronic wounds (16, 17). Their success was multifactorial (18). Firstly, chronic wounds are a common skin condition, especially in the aging population and immunocompromised patients. Secondly, the treatment modalities for this disease are still quite limited. Lastly, the antimicrobial and anti-inflammatory features of CAP make it an ideal tool for a chronically colonised wound.

Since then, several reviews have been published on this topic, but each with pros and cons (19–24). Some are oversimplified in terms of disease spectrum, others failed to balance the clinical findings and scientific interpretation. Nonetheless, this is the first relevant review that is comprehensive, up-to-date, disease-specific, and bridging the gap between biotechnology and clinical practice. Literatures from only the last decade were extracted from PubMed and presented here in three aspects: optimising intact skin, treating skin disease, and improving drug absorbance into skin.

Human skin is a unique structure, with a surface area of 2 m² and makes up for up to 20% of the total body weight. Skin is multifunctional as it provides protective barrier, acts as sensory receptor, transports nutrients and metabolites, helps regulate body temperature, and exercises immunological activity. Skin is composed of three layers, from superficial to deep: the epidermis, dermis, and hypodermis ( Figure 2 ).

The epidermis is further divided into five sublayers: stratum corneum, stratum lucidum, stratum granulosum, stratum spinosum, and stratum basale. The basic cell types of epidermis include keratinocytes, melanocytes, and Langerhans cells. Keratinocytes are of ectodermal origin and have the specialized function of producing keratin. Melanocytes derive from the neural crest and synthesize melanin. Langerhans cells are dendritic cells and can recognize, process and present antigens to lymphocytes during hypersensitivity reaction.

The dermis is connected to the epidermis through the basement membrane zone. The dermis is divided into two areas: papillary region (superficial area) and reticular region (deep area). It contains adnexal structures, such as sweat glands, hair follicles, sebaceous glands, and nails. The principal component of the dermis is collagen, particularly type I collagen, which is the major stress-resistant substance of the skin.

The hypodermis, or subcutaneous tissue, consisting of loose connective tissue and adipose tissue, attaches skin to the underlying muscles and bones. Here, the main cell types are fibroblasts, macrophages, and lipocytes. The hypodermis also contains larger blood vessels and nerves which supply the skin.

There have been several in vitro studies in the last few years focusing on the effect of CAP exclusively on non-diseased skin cells ( Table 1 ). These projects mostly used HaCaT human immortalized keratinocytes as the cellular model. Although there were some variations in plasma-induced cellular changes, it was found that the respective change at the cellular level was also reflected correspondingly at the protein level and gene level.

In a relatively early study, Arndt et al. used the MicroPlaSter β plasma torch system on primary human skin keratinocytes (30). Their main novel finding was that antimicrobial peptides of the β-defensin family were upregulated after CAP treatment. Defensins are small cysteine-rich cationic proteins that are active against bacteria, fungi and viruses. They could be found in inflammatory skin conditions such as atopic dermatitis. CAP also induced gene expression of key regulators for inflammation and wound healing, such as interleukin 8 (IL-8) and transforming growth factor beta (TGF-β). However, the proliferation, migration and apoptosis of keratinocytes were not altered by CAP. Subsequent studies using HaCaT cells found different results, as cell proliferation, motility and apoptosis could all be changed using CAP treatment (25–28).

Choi et al. used a low-frequency argon plasma device based on the dielectric barrier discharge (DBD) technology (25). At the cellular level, keratinocyte proliferation was stimulated by CAP, as the transition from G1 cell cycle phase to S and G2 phases was quickened. At the protein level, CAP not only dispersed E-cadherin-mediated cell-to-cell interactions but also translocated β-catenin from the cytosol to the nucleus. At the gene level, CAP treatment increased the expression and transcription of c-MYC and cyclin D1. Therefore, these results of enhanced epidermal cell growth demonstrated that CAP might be used a novel skin regenerating apparatus.

On the contrary, a series of studies conducted by Schmidt et al. using kINPen 09 atmospheric argon plasma jet showed suppressive results instead. Firstly, reduction in cell proliferation, transient increase in cell migration, and secretion of immunomodulatory signal proteins were found in human keratinocytes after direct CAP exposure (26). The above changes were centred around the p53 regulatory axis, with upstream phosphorylation of ATM and ATR and downstream activation of checkpoint kinases Chk and mitogen-activated protein (MAP) kinases. Thus, plasma-induced proapoptotic, proinflammatory, and pro-survival effects were confirmed. Secondly, periodic, long-term, and indirect treatment using plasma-activated medium (PAM) repressed keratinocyte motility and enlarged cell size (27). The authors discovered differential expression of 260 genes involved in inflammation and redox homeostasis using transcriptomic microarray. The protein products of these deregulated genes include various cytokines, growth factors, antioxidant enzymes, and apoptotic signalling targets. Lastly, the same research group changed reactive species generation of the plasma effluent by modulating the ambient gas composition from pure nitrogen to pure oxygen (28). The oxygen-shielding plasma provided stronger apoptotic effect than the nitrogen counterpart and induced keratinocyte response more efficiently. Gene expression analysis revealed induction of signalling and communication proteins such as immunomodulatory interleukin 6 (IL-6) and several antioxidative molecules.

In summary, the inconsistency in plasma-induced cellular change in healthy skin cells, such as cell proliferation, cell migration, and cell apoptosis are multifactorial. CAP-generating technology (e.g., DBD or others), treatment modality (e.g., direct jet or indirect PAM), plasma parameters (e.g., duration and gas composition), and target skin cell type (e.g., keratinocytes vs. fibroblasts, human vs. murine) all could be contributing factors. More dedicated studies are required to expand our understanding on the effect of CAP on healthy skin cells.

The above CAP-induced cellular effects have been further examined at the histological level. Additional architectural information was obtained using various histological staining methods including haematoxylin and eosin (H&E) staining, Masson’s trichrome staining, and immunohistochemical (IHC) analysis. The two consecutive studies published by Choi et al. revealed that short period of CAP treatment on HRM2 mice caused significant increases in epidermal thickness and dermal collagen density (25, 31). Repeated CAP treatment not only caused epidermal expansion by activating β-catenin in the epidermal cells, but also accelerated dermal remodelling. On the other hand, plasma increased the skin tissue expression of various growth factors, such as TGF, vascular endothelial growth factor (VEGF), granulocyte macrophage-colony stimulating factor (GM-CSF), and epidermal growth factor (EGF). Interestingly, Hasse et al. brought more clinical value as they used human skin biopsies from nine patients for plasma treatment (32). Unlike what was shown previously where plasma-generated species only penetrated the first few layers of the epidermis (33), the pro-proliferative effect of CAP in the current work occurred as deep as the stratum basale, while the epidermal integrity and keratin expression remained unchanged. Collectively, these in vivo studies support CAP as an innovative tool with anti-aging potential for the skin.

In general, CAP can improve cutaneous microcirculation of intact skin. In 2016, Kisch et al. organized a controlled, prospective cohort study including 20 healthy patients using skin at the radial forearm (34). Their DBD CAP device increased cutaneous tissue oxygen saturation by 24% and maintained effective for at least 8 min. In addition, the cutaneous capillary blood flow increased by 73% and remained fast for 11 min. Furthermore, even more prominent elevation of these parameters was found in patients with lower body mass index (BMI).

In comparison, Borchardt et al. conducted a similar study with a smaller number of patients, but assessing lot more circulation parameters for a longer monitoring period (35). Firstly, tissue oxygen saturation and blood flow both improved in a plasma duration-dependent manner. Secondly, skin pH decreased by 0.3 after CAP treatment, whereas skin temperature and moisture were not altered. Most importantly, CAP-related enhancement of skin microcirculation was found to be specific to the plasma treatment and not a result of the applied mechanical pressure.

In order to study the influence of CAP on vascularization-involved molecules in skin-related cells, Arndt et al. performed in vitro and in vivo experiments on a microscopic scale using the MicroPlaSter β plasma torch system (36). CAP significantly activated the expression of artemin, EGF, endocrine-derived VEGF (EG-VEGF), endothelin-1, basic fibroblast growth factor (bFGF), IL-8 in human epidermal keratinocytes, angiogenin, endostatin, monocyte chemoattractant protein (MCP)-1, matrix metalloproteinase (MMP)-9, tissue inhibitor matrix metalloproteinase (TIMP)-1, and VEGF in human dermal fibroblasts, as well as angiopoietin (Ang)-2, angiostatin, amphiregulin, endostatin, bFGF and angiogenic receptors in human umbilical vein endothelial cells (HUVECs). It was concluded that CAP modulates angiogenesis-related factors in an autocrine and paracrine mode.

The lipids of the stratum corneum is one of the most important components of the skin barrier. Marschewski’s group was one of the first to use X-ray photoelectron spectroscopy (XPS) to analyse CAP-induced changes in the physiological skin lipid composition (37). Using stripped off skin lipids from human forearms, the authors revealed that the total carbon amount reduced from 84.4% to 76.7, oxygen increased from 10.8% to 16.5%, and nitrogen marginally increased from 4.8% to 6.8%. These chemical changes were due to reduced C-C bonds and increased C-O, C=O, C-N, and N-C-O bonds. This proof-of-principle investigation was consolidated by Striesow’s in vitro experiment in which collected human forehead lipids were treated using an argon plasma jet (38). The direct-infusion high-resolution tandem mass spectroscopy (DI-MS²) and liquid chromatography-tandem mass spectroscopy (RP-LC/MS²) detected minimal CAP-driven oxidation of triacyclglycerols, ceramides, and cholesteryl esters. In functional term, epidermal lipid overlay could be well protected, and moderate CAP treatment would result in limited negative consequence in the dermal tissue. In addition, Schmidt et al. conducted an in vivo study on murine skin and discovered oxidative modification in the relative abundance of lipid classes using reversed-phase liquid chromatography/mass spectroscopy (39). Finally, Kartaschew et al. used infrared and Raman vibrational microspectroscopy to chemically analyse the plasma-induced change in skin components, such as keratin and lanolin which resembles human sebum (40). The authors suggested that the resultant acidic and NO-containing functional groups could be the source of an antibacterial and regenerative environment in the stratum corneum where plasma interacted with.

Although the aforementioned investigational hardware, such as various types of mass spectroscopy, have greatly facilitated the study of the effect of CAP on skin chemistry, limitations of the analytical software’s ability to identify unexpected oxidized lipids could lead to an underestimation of CAP’s impact on skin lipids (38), suggesting a need for software advancement.

Most bacterial skin infections are caused by gram-positive organisms, such as Staphylococci and Streptococci. Since the antibacterial effect of CAP has already been discussed in ‘Cutaneous wound healing’ (section 3.1.1.) and other reviews, treatment of bacterial skin disease will not be elaborated here. More focus is placed on CAP therapy for fungal and viral skin diseases.

Demodicosis in humans is usually caused by Demodex folliculorum mites, which may be a cause or exacerbating factor in papulopustular rosacea. Rosacea is a chronic and disfiguring condition that typically affects the midfacial skin. Daeschlein et al. successfully used CAP in vitro as a non-antimicrobial treatment to inactivate Demodex folliculorum isolated from a patient (83). CAP’s antiparasitic efficacy could suggest an alternative to Rosacea treatment.

Pediculosis is an infestation of lice including three types, namely pediculosis capitis (head louse infestation), pediculosis corporis (body louse infestation), and pediculosis pubis (pubic louse infestation). Ten Bosch et al. developed a comb-like CAP device designed as a physical remedy for pediculosis capitis (82). Their results showed that a single stroke of plasma over a hair strand led to high mortality rates of P. humanus humanus and parasitic eggs. The safety of this novel insecticide-free option was also verified in three aspects with satisfaction, e.g., ozone concentration, UV emission, and patient leakage current.

Skin cancer include three main types, i.e., basal cell carcinoma (BCC), squamous cell carcinoma (SCC), and melanoma, which is in decreasing order of prevalence and increasing order of aggressiveness. One of the foundations for plasma oncology is CAP’s ability to selectively kill cancer cells over surrounding healthy cells (112). Most of recent relevant work are cellular and animal studies.

In an in vitro cellular study, Lee et al. manipulated cell cycle progression of human hair follicle dermal papilla cells (DPCs) using CAP-activated medium (105). Human DPCs are mesenchymal cells isolated from the hair papilla of healthy scalp hair follicles. In the adult hair follicle, hair papilla plays a key role in controlling hair production and hair growth (113). PAM treatment at an appropriate concentration and duration was found to arrest cell proliferation and cell cycle of DPCs, which could gradually recover.

In an in vivo animal study using Wistar rats, Babossalam et al. achieved skin rejuvenation using a novel pulsed nitrogen plasma touch (107). A significant increase in epidermal thickness, fibroblast proliferation, and skin collagenesis were discovered. Most importantly, CAP also increased the diameter of primary and secondary hair follicles in the treated skin.

In a human trial including 14 patients with androgenic alopecia, Khan et al. tested long-term CAP treatment in the form of PAM (106). The theoretical bases for this study were CAP’s ability to induce stem cell differentiation in various cell types (3), and its inconsistency in terms of depth of penetration, especially on thick human scalp skin. The six-month periodic PAM washing was well tolerated, and most patients and clinicians reported subjective and qualitative improvement.