Introduction
Although skin absorption of organic solvents in the liquid phase is well recognised there has been less data on absorption of solvent vapours through the skin. There have been some studies looking at dermal absorption of some solvents (either liquid or as vapours) in human volunteers (Brooke et al, 1998; Johanson and Boman, 1991; Corley et al, 1997, Kezic et al, 1997, Mraz and Nohova, 1992). However, in the UK at least, human volunteer studies are restricted to substances with health-based Occupational Exposure Limits. In addition, human volunteer studies require ethics approval for each study, are expensive and can only use a small number of volunteers.

The rise of in-vitro techniques for studying percutaneous penetration allows the study of chemicals that would be unethical to study in volunteers and offers the potential to use more 'individuals' (donors) allowing a better understanding of inter-individual variation. However, there are differences in response between in vivo and in vitro systems and these need to be studied before using in vitro techniques as a replacement for human volunteer studies. Although in vitro systems will never replicate the entire in vivo system, they do provide an opportunity to study specific skin related issues.

We have conducted a number of human volunteer studies looking at the dermal absorption of solvent vapours and have replicated these investigations using an in vitro percutaneous absorption cell (IVPAC).

Methods
The solvent vapours under study were 1-methoxy-2-propanol, 2-butoxyethanol, 2-butanone (MEK), and xylene.

In vivo studies
Exposures were performed at the Health and Safety Laboratory Controlled Atmosphere Facility, a purpose built room of approximately 8 m3 volume, as previously described (Brooke et al, 1998). Volunteers were exposed on 2 occasions - once as a 'whole body' exposure and once as a 'skin only' exposure. For the 'skin only' exposure, the volunteers wore air-fed half-masks so that the inhalation route was excluded as a source of uptake. The body burden of each exposure was determined by biological monitoring and the body burden of the 'skin only' exposure was then expressed as a percentage of that of the 'whole body' exposure in order to determine the extent of dermal absorption. Volunteers acted as their own control.

Absorption of 1-methoxy-2-propanol, 2-butoxyethanol, 2-butanone and xylene was determined by post-exposure blood sampling, measuring 1-methoxy-2-propanol, 2-butoxyacetic acid, 2-butanone and xylene respectively. Butoxyethanol was not measured as a marker itself as there has been some controversy over blood butoxyethanol measurements (Johanson and Boman, 1991; Corley et al, 1997).

In vitro studies
Dermatomed human skin samples (approx 700 micrometers thick) were mounted in IVPAC diffusion cells (Griffin et al, 2000) and maintained at 32 C in a heated manifold. The skin surface (donor chamber) was exposed to a constant flow of the vapour (20 L/min) in each diffusion cell, and the undersurface of the skin (receptor chamber) was exposed to a constant flow of a blood surrogate, the receptor fluid (10% w/v bovine serum albumin, 1.5 ml/hr). The receptor chamber contents were magnetically stirred. Vapour penetrating the skin was absorbed into the receptor fluid and eluted into cooled vials (on cardice) held in a fraction collector. Receptor fluid samples were collected at 30 minute intervals and analysed by GC-MS.

Results
Table 1 summarises the exposure conditions for all the studies.


Table 1. Concentration (ppm) and length (hours) of exposure in the various studies.

Table 2. Comparison of results from in vivo and in vitro studies.

Figure 1 compares in vivo and in vitro data. When the scales are adjusted to allow overlay, it can be seen that the in vivo and in vitro data are in broad agreement as to the extent of absorption of solvent vapours. As expected 2-butoxyethanol and methoxy-2-propanol are most readily absorbed through the skin although there is not the large discrepancy in absorption that might be expected from the predicted permeation coefficients (Kp) which range from 0.7 cm/hr (xylene) to 27.3 cm/hr for butoxyethanol.

Figure 1. Comparison of in vivo and in vitro data.

Conclusions
The in vitro data compare well with the in vivo data in ranking the extent of dermal absorption of solvent vapours and to some extent in the relative magnitude of the absorption. The glycol ethers (methoxy-2-propanol and butoxyethanol) are, as expected, most readily absorbed both in vivo and in vitro although both predict butoxyethanol to have the greatest dermal absorption potential. This is in contrast to Dugard et al (1984) who reported that the in vitro absorption rate of methoxy-2-propanol was nearly 6x that of butoxyethanol (1.17 mg.cm-2.h-1 compared to 0.198 mg.cm-2.h-1) for absorption of the neat liquid solvents, although there was considerable variation in their results. This contradiction may indicate a difference in behaviour of solvents in the liquid and vapour phase.

Our studies have shown that in vitro absorption data may be useful in ranking solvents as to their dermal absorption potential and may allow estimates of potential skin absorption of chemicals that cannot ethically be studied in vivo, such as carcinogens and sensitisers. Further work is needed on developing relationships between an in vitro dermal absorption rate and an in vivo body burden.

References
Brooke I, Cocker J, Delic JI, Payne M, Jones K, Gregg NC, Dyne D. (1998) Dermal uptake of solvents from the vapour phase: an experimental study in humans, Ann Occup Hyg. 42(8):531-40.

Corley RA, Markham DA, Banks C, Delorme P, Masterman A, Houle JM. (1997) Physiologically based pharmacokinetics and the dermal absorption of 2-butoxyethanol vapor by humans, Fundam Appl Toxicol. 39(2):120-30.

Dugard PH, Walker M, Mawdsley SJ, Scott RC. (1984) Absorption of some glycol ethers through human skin in vitro, Environ Health Perspect. 57:193-7.

Griffin P, Payne M, Mason H, Freedlander E, Curran AD, Cocker J. (2000) The in vitro percutaneous penetration of chlorpyrifos, Hum Exp Toxicol. 19(2):104-7.

Johanson G, Boman A. (1991) Percutaneous absorption of 2-butoxyethanol vapour in human subjects, Br J Ind Med. 48(11):788-92.

Kezic S, Mahieu K, Monster AC, de Wolff FA. (1997) Dermal absorption of vaporous and liquid 2-methoxyethanol and 2-ethoxyethanol in volunteers, Occup Environ Med. 54(1):38-43.

Mraz J, Nohova H. (1992) Absorption, metabolism and elimination of N,N-dimethylformamide in humans, Int Arch Occup Environ Health. 64(2):85-92.