We initially failed to confirm a case of polycystic ovary syndrome (PCOS) because underestimation of LH concentrations due to a variant form of this hormone resulted in a misleadingly low LH/FSH ratio. A 26-year-old woman presented to our hospital with infertility. Given the presence of bilateral polycystic ovaries, oligomenorrhea and hirsutism, PCOS was suspected, but a normal LH/FSH ratio as measured by RIA led to diagnostic problems. When we remeasured LH and FSH using a chemical luminescence enzyme immunoassay (CLEIA), the ratio of the LH concentration measured by RIA to that measured by CLEIA was 0.29, and the ratio of LH to FSH measured by CLEIA was 3.3 compared with 0.81 measured by RIA. We then diagnosed PCOS.