In an attempt to differentiate an allergic patch test response from an irritant response, we evaluated by flow cytometry the percentages of various epidermal cell populations isolated from allergen and irritant-treated patch test sites. Nine allergic individuals were patch tested with various allergens (Rhus, dinitrochlorobenzene [DNCB], or nickel chloride) and a vehicle control for 48 h. Eight additional individuals were patch tested with irritating chemicals (sodium lauryl sulfate or nonanoic acid) and with a vehicle control for 48 h. Epidermal cells, isolated from suction blisters, were double labeled for CD1/HLA-DR, CD3/HLA-DR, or CD36/HLA-DR cell surface markers and analyzed by flow cytometry to determine the percentage of various cell populations. A mean increase of 0.91 +/- 0.3 in the percentage of DR+CD1+ Langerhans cells over the vehicle control patch test site was detected in allergen-positive patch test sites in allergic individuals, whereas a decrease of 0.19 +/- 0.2 in the percentage of DR+CD1+ Langerhans cells from the vehicle control patch test site was detected in irritant-treated patch test sites. Epidermal cells from allergen-positive patch test sites also exhibited an increase of 5.2 +/- 1.8 in percentage of DR+CD1- cells over the vehicle control patch test site compared to an increase change of 0.8 +/- 0.4 in epidermal cells isolated from irritant-treated patch test sites. We also found that DR+ cells that lacked the CD1 determinant expressed the macrophage/monocyte antigen CD36 (OKM5). Finally, a 2.3 +/- 0.8 increase in the percentage of DR-CD3+ cells over the vehicle control patch test site was observed in allergen-positive patch test sites compared to an increase of 0.2 +/- 0.2 observed in irritant-treated patch test sites. These results demonstrate a significant increase in DR+CD1+, DR+CD1-CD36+, and DR-CD3+ epidermal cells in allergen-positive patch test sites compared to irritant patch test sites.