Graves and colleagues [1] have shown that TNFα concentrations were elevated in the wounds of mice used as a model of diabetes compared with normoglycaemic littermates and that TNFα drove fibroblast apoptosis, hence impairing wound healing. In their study, anti-TNF treatment increased fibroblast density and improved wound healing. Of course, the usual caveat applies to such animal studies (the authors used streptozotocin-induced and db/db models of type 1 and type 2 diabetes, respectively) regarding whether they are relevant to human disease, but the results of Siqueira et al. [1] are in close agreement with observations we have made in a human model of dermal wound healing employing cantharidin skin blisters.Cantharidin is a blistering agent used in the treatment of molluscum contagiosum and warts [2]. It is a protein phosphatase 1 and 2 alpha inhibitor that causes acantholysis [2]. When applied to healthy skin it can model the proinflammatory and resolving phases of wound healing, depending on the duration of blister: at 16 h blister fluid contains proinflammatory mediators, such as TNFα, IL-8 (also known as CXCL8) and monocyte chemoattractant protein-1 (MCP-1; also known as CCL2), but at 40 h these have substantially diminished to be replaced by wound healing factors, such as TGFβ [3,4]. Early blisters are infiltrated with a predominance of polymorphonuclear leucocytes and monocytes, whereas later skin blisters show evidence of neutrophil apoptosis and differentiation of monocytes into macrophages, characteristics of inflammatory resolution.To investigate whether the balance of inflammation and resolution was disturbed in persons with diabetes, nine patients with type 2 diabetes mellitus were recruited at the Hammersmith Hospital with informed consent and institutional review board approval, in accordance with the principles of the Declaration of Helsinki as revised in 2000. Twelve healthy individuals were recruited as a control group. Two skin blisters were created on the volar aspect of the forearm by topical application of cantharidin (Cantharone, Dormer Laboratories, ONT, Canada) at 0.1% (vol./vol.) in acetone. One skin blister was harvested at 16 h and one blister at 40 h of duration. Differential leucocyte counts in blister fluid were performed by Kimura's stain and neutrophil apoptosis was determined by flow cytometry, monitoring loss of CD16 staining as an early marker of apoptosis [5,6]. It was not possible within the ethical constraints of the protocol to investigate keratinocyte or fibroblast apoptotic or proliferative responses. Concentrations of TNFα in blister fluid were determined by ELISA.