Imrecoxib is a typical cyclooxygenase-2 inhibitor and the benzylic carbon motif is its major site of oxidative metabolism, producing a hydroxymethyl metabolite (M1) and a carboxylic acid metabolite (M2). The plasma exposure of M2 is four times higher than those of both M0 and M1 in humans. However, this metabolite is rarely formed in in vitro experiments. Therefore, this study aims to investigate the formation mechanism of M2 and to further elucidate the reason for the discrepancy between in vitro and in vivo metabolic data. By employing human hepatocytes, human liver microsomes (HLMs), human liver cytosols (HLCs), recombinant enzymes, and selective enzyme inhibitors, the metabolic map of imrecoxib was elaborated as follows: the parent drug was initially hydroxylated to form M1 in HLMs, mainly mediated by CYP3A4 and CYP2D6, and to subsequently form aldehyde imrecoxib (M-CHO) in HLMs and HLCs. The latter process is the rate-limiting step in generating the end-product M2. In further M-CHO metabolism, two opposite reactions (namely, rapid oxidation catalyzed by CYP3A4, CYP2D6, and cytosolic aldehyde oxidase to form M2 versus reduction to regenerate M1 mediated by NADPH-dependent reductases in HLMs and HLCs, such as cytochrome P450 reductase) led to marked underestimation of the M2 amount in static in vitro incubations. The findings provided a possible explanation for the difference between in vitro and in vivo metabolism of imrecoxib, suggesting that the effect of competitive reduction on the static oxidation metabolism in in vitro metabolic experiments should be considered.