Effect of selective versus nonselective cyclooxygenase inhibitors on gastric ulceration scores and intestinal inflammation in horses

Abstract

AbstractObjectiveTo determine whether a cyclooxygenase (COX)‐2 selective nonsteroidal anti‐inflammatory drug (NSAID) would reduce gastric ulceration and gastrointestinal (GI) inflammation compared with a non‐COX selective NSAID.Study designRandomized block design.AnimalsTwenty‐five healthy adult horses.MethodsHorses were randomly assigned to receive placebo (n = 5), phenylbutazone (n = 10), or firocoxib (n = 10) administered daily for 10 days. Gastroscopy was performed on days 0 and 10, and both squamous and glandular ulcers were scored according to established scoring criteria. Fecal samples were collected on days 0, 10, and 20 to test for fecal myeloperoxidase (MPO) concentration by enzyme‐linked immunosorbent assay.ResultsBoth classes of NSAID induced GI injury as determined by gastric ulceration scores and fecal MPO. Glandular gastric ulceration scores and fecal MPO concentrations were higher in horses treated with phenylbutazone at day 10 (P < .001 and P = .0018, respectively). Increases in fecal MPO were significantly decreased 10 days following cessation of treatment for firocoxib but remained greater than baseline for the phenylbutazone group.ConclusionAlthough both classes of NSAID induced gastric ulceration, the COX‐2 selective NSAID firocoxib induced less severe glandular ulceration. Although there were increases in fecal MPO in both groups after 10 days of treatment, this increase was significant only in horses receiving the nonselective COX inhibitor phenylbutazone.Clinical significanceThese findings suggest that both classes of NSAID induce GI injury in horses; however, at the dosages used in this study, the COX‐2 selective NSAID firocoxib resulted in less severe injury.