Dermal (percutaneous, skin) absorption is a global term that describes the transport of chemicals from the outer surface of the skin both into the skin and into the systemic circulation. This Environmental Health Criteria document presents an overview of dermal absorption and its application to the risk assessment of chemicals. In addition, it presents and discusses current topics of interest in the field of dermal absorption. Dermal absorption can occur from occupational, environmental, or consumer skin exposure to chemicals, cosmetics, and pharmaceutical products. The skin is a complex organ and a living membrane. The functions of the skin include protection, regulation of body temperature and water loss, and defence and repair. The skin is composed of an outer region, the epidermis, and an inner region, the dermis. The epidermis consists of various cell layers, the outermost layer, the stratum corneum or horny layer, functioning as the main barrier to the entry of extraneous chemicals. The viable epidermis can metabolize chemicals that pass through the stratum corneum. The dermis provides physiological support for the avascular epidermis and is the locus of blood vessels, sensory nerves, and lymphatics in the skin. The skin also contains appendages, such as hair follicles, sweat glands, and sebaceous glands, which originate in the subpapillary dermis. There is considerable variability in the measurement of skin permeability. There can be major differences in permeability between species. Little is known about variation due to age, although the skin structure does change with age; however, sex and ethnic background do not seem to be sources of variation in permeability. Percutaneous absorption is also dependent on the anatomical site, on the skin condition, and on the hydration state of the skin. Factors influencing percutaneous absorption through the skin include 1) physicochemical properties of the test compound, 2) physicochemical and other properties of the vehicle in which the test compound is dissolved, 3) interactions between the test compound or vehicle and the skin, 4) skin properties and metabolism, and 5) factors inherent to the test system used for measurement - for example, dose and volume of test substance, occlusion or non-occlusion of test area, in vitro or in vivo test systems, and duration of exposure. Theoretical equations and models have been developed to describe the transport of a diffusing chemical through the skin. Typically, the steady-state flux (Jss) and the permeability coefficient (Kp) are the main parameters assessed from in vitro experiments in which the donor concentration of the penetrant is maintained at constant (infinite) dose conditions. The estimation of maximum flux, time for maximum flux, lag time, residual (reservoir) amounts retained in the stratum corneum, and mass balance under "real" application conditions is now recognized to be of prime importance in exposure estimations. Measurement of metabolism of a chemical in contact with skin may be important in both efficacy and safety evaluations. Some chemicals can be significantly metabolized during dermal absorption, which may result in either inactive or active metabolites. The measurement of this metabolism may therefore be important in an appropriate safety evaluation. Toxic chemicals such as benzo[a]pyrene have been shown to be activated in skin, whereas other chemicals may undergo hydrolysis and/or conjugation reactions in the skin, resulting in a decrease in the availability of those chemicals to the body. In general, the viability of skin can be maintained in an in vitro diffusion cell by using fresh skin and a physiological buffer. It is recommended that this viability be verified through measuring the activity of relevant metabolizing enzymes. The permeability properties of the stratum corneum are, for the most part, unchanged after its removal from the body. As a consequence, a good correlation exists between measurements derived from both in vivo and in vitro skin diffusion experiments with the same chemicals (at least for hydrophilic compounds). In vitro experiments are an appropriate surrogate for in vivo studies and offer a number of advantages over whole-animal or human volunteer experiments. In vitro methods measure the diffusion of chemicals into and across skin to a fluid reservoir and can utilize non-viable skin to measure diffusion only or fresh, metabolically active skin to simultaneously measure diffusion and skin metabolism. Test Guideline 428 of the Organisation for Economic Co-operation and Development (OECD) encourages harmonization of methodology. Experimental factors affecting dermal absorption in vitro, in addition to those mentioned above, include the thickness of skin sample, variations in temperature of the test system, and composition of the receptor fluid. Static or flow-through in vitro diffusion cells can be used. Additional techniques, requiring further refinements, include tape stripping and the use of artificial or reconstituted skin. In vivo methods allow the determination of the extent of cutaneous uptake as well as systemic absorption of the test substance. The main advantage of performing an in vivo study rather than an in vitro study is that it uses a physiologically and metabolically intact system. In vivo dermal penetration studies are carried out in laboratory animals, usually rodents, and in human volunteers. In vivo dermal penetration studies in human volunteers have been widely used for human pharmaceuticals and, to a more limited extent, for other chemicals. In vivo studies in humans are the gold standard. The conduct of any in vivo study has ethical issues. The main disadvantage in the use of laboratory animals is that they have different skin permeability and systemic disposition compared with humans. The results of human volunteer studies have shown that occupational exposure to liquids (such as solvents) can result in considerable dermal absorption. Skin uptake from vapours may be an important contributor to the total uptake for some volatile substances, such as the glycol ethers. In vitro dermal absorption studies are increasingly being submitted for registration purposes for industrial chemical, cosmetic, and crop protection products. There are many published studies that compare in vitro and in vivo results in laboratory animals and humans. Properly conducted in vitro studies that follow the OECD test guidelines have demonstrated that the in vitro approach can provide good prediction of in vivo dermal absorption. Over the decades, a large number of data have been generated on the percutaneous penetration of a wide range of chemicals, pesticides, cosmetics, and pharmaceuticals. Studies have included work on human volunteers, in vivo studies using animal models, in vitro studies on excised human, rodent, pig, guinea-pig, etc., skin, and, more recently, in vitro studies on synthetic skin. There have been many attempts over the last 50 years to predict the rate and extent of dermal absorption and so reduce the need for in vitro and in vivo testing. This need is even greater in response to increasing ethical difficulties with respect to human and laboratory animal experiments as well as the legislatively imposed economic and time considerations - particularly in the risk assessment of industrial chemicals. Quantitative structure-permeability relationships (QSPeRs) are statistically derived relationships between the steady-state flux of a compound and various physicochemical descriptors and/or structural properties of the molecule. Efficacy and safety considerations also recognize quantitative structure-activity relationships (QSARs) in irritation, skin sensitization, metabolism, chemical effects, and clearance. QSARs are therefore involved at a number of levels in chemical safety. Mathematical models have been used to simulate the dynamics of the partition, diffusion, metabolism, and other processes involved in dermal absorption and can lead to the prediction of the extent and rate of chemical permeation through the skin. Mathematical modelling plays a key role in linking permeability coefficient and flux data obtained from tests under steady-state conditions (i.e. infinite dose) to absorption estimates for finite dose applications that are more typical of occupational exposure (i.e. non-steady-state conditions). In risk assessment, the initial estimate for dermal absorption is usually obtained by the use of a tiered approach, where the greatest safety margin is defined by the worst case and more refined estimates better define the real margin. Hence, as a first step, 100% absorption is assumed when no data are available. In the second step, a more realistic estimate of the extent of dermal absorption is provided by a consideration of the physicochemical properties of the chemical and the vehicle. The third step is a consideration of any experimental in vitro and in vivo dermal absorption data. If, at the end of these steps, an unacceptable risk is calculated, the risk assessment is best refined by means of actual exposure data. In the last few years, partly due to regulatory pressures, there have been several initiatives to accelerate progress in the fields of international harmonization of methodology and protocols, culminating in the publication of the OECD test guidelines for skin absorption studies in 2004. This international collaboration includes projects such as an international validation study involving 18 laboratories, the European Evaluations and Predictions of Dermal Absorption of Toxic Chemicals (EDETOX) project, and projects sponsored by industry, as well as conferences, such as the Perspectives in Percutaneous Penetration (formerly the Prediction of Percutaneous Penetration, or PPP) and Gordon Research conferences.