Clinical utility of phosphodiesterase 4 (PDE4) inhibitors as anti-inflammatory agents has, to date, been limited by adverse effects including nausea and emesis, making accurate assessment of emetic versus anti-inflammatory potencies critical to the development of inhibitors with improved therapeutic indices. In the present study we determined the in vitro and in vivo anti-inflammatory potencies of the first-generation PDE4 inhibitor, rolipram, the second-generation inhibitors, roflumilast and cilomilast, and a novel third generation inhibitor, 1-ethyl-5-{5-[(4-methyl-1-piperazinyl)methyl]-1,3,4-oxadiazol-2-yl}-N-(tetrahydro-2H-pyran-4-yl)-1H-pyrazolo[3,4-b]pyridin-4-amine (EPPA-1). The rank-order potency against lipopolysaccharide (LPS)-induced tumor necrosis factor-alpha production by human peripheral blood mononuclear cells was roflumilast (IC(50) = 5 nM) > EPPA-1 (38) > rolipram (269) > cilomilast (389), and against LPS-induced pulmonary neutrophilia in the rat was EPPA-1 (D(50) = 0.042 mg/kg) > roflumilast (0.24) > rolipram (3.34) > cilomilast (4.54). Pica, the consumption of non-nutritive substances in response to gastrointestinal stress, was used as a surrogate measure for emesis, giving a rank-order potency of rolipram (D(50) = 0.495 mg/kg) > roflumilast (1.6) > cilomilast (6.4) > EPPA-1 (24.3). The low and high emetogenic activities of EPPA-1 and rolipram, respectively, detected in the pica model were confirmed in a second surrogate model of emesis, reversal of alpha(2)-adrenoceptor-mediated anesthesia in the mouse. The rank order of therapeutic indices derived in the rat [(pica D(50))/(neutrophilia D(50))] was EPPA-1 (578) > roflumilast (6.4) > cilomilast (1.4) > rolipram (0.15), consistent with the rank order derived in the ferret [(emesis D(50))/(neutrophilia D(50))]. These data validate rat pica feeding as a surrogate for PDE4 inhibitor-induced emesis in higher species, and identify EPPA-1 as a novel PDE4 inhibitor with an improved therapeutic index.