Harvell, J. D., V. Gordon, and H. I. Maibach. Use of the Skintex High Sensitivity Assay for Predictive Assessments of Cutaneous Fatty Acid Irritancy in Man. In Vitro Toxicol. 1993. 6(2): 91-96.

propanoic acid - 79-09-4; pentanoic acid - 109-52-4; hexanoic acid - 142-62-1; octanoic acid - 124-07-2; nonanoic acid - 112-05-0; decanoic acid - 334-48-5; undecyclic acid - 112-37-8; myristic acid - 544-63-8; hexadecanoic acid - 57-10-3; stearic acid - 57-11-4; oleic acid - 112-80-1; linoleic acid - 60-33-3; linolenic acid - 463-40-1; n-propanol - 71-23-8; tridecanoic acid - 638-53-9; pentadecanoic acid - 1002-84-2; heptadecanoic acid - 506-12-7; elaidic acid - 112-79-8

We have tested the irritancy of fatty acids of chain lengths C3-C18 in an in vitro dermal irritancy assay - the Skintex High Sensitivity Assay. The results of a modified in vitro protocol (which more closely mimicked the in vivo protocol) were compared to previously published human data (Stillman, et al., 1975). The previously published in vivo data set represented an example of cumulative (chronic) irritancy (i.e. several repeat applications were needed to produce an irritant effect in vivo). Comparing the in vivo results to the in vitro results, the following parameters for this data set were calculated: sensitivity 83%, specificity 50%, positive predictive value 77%, and negative predictive value 60%. These moderate results may reflect either (1) a difficulty of the Skintex system to accurately predict the results of cumulative irritancy, or (2) the difficulty in defining a human irritant/nonirritant cutoff point, which influences these measures. For example, when the human in vivo irritancy cutoff is designated a 2, instead of a 1, the sensitivity becomes 100%, specificity 50%, positive predictive value 62%, and negative predictive value 100%.