Polycystic ovary syndrome (PCOS) is the most common endocrine disorder of the reproductive-aged women with considerable metabolic and reproductive morbidity. There has been an ongoing debate regarding definition and diagnostic criteria of PCOS since its first description by Stein and Leventhal in 1935. The most widely used diagnostic criteria (so-called 1990 NIH criteria) define PCOS as the presence of hyperandrogenism and oligo-ovulation after exclusion of the other known disorders. Alternatively, a recently convened international workshop sponsored by European Society of Human Reproduction and Embryology (ESHRE)/American Society for Reproductive Medicine (ASRM) suggested that PCOS should be defined by the presence of two of the following three criteria: (1) oligo- or anovulation; (2) clinical and/or biochemical signs of hyperandrogenism; and (3) polycystic ovaries (PCO). The Rotterdam conference provided clear guidelines for determination of clinical, hormonal, and biochemical features of PCOS, and acknowledged the limitations and difficulties in assessment of clinical hyperandrogenism, hyperandrogenemia, and PCO. Essentially, all patients diagnosed by 1990 NIH criteria would also be defined as having PCOS according to 2003 Rotterdam criteria. However, Rotterdam conference introduced two new phenotypes: (1) hyperandrogenism with PCO and normal ovulation, and (2) PCO and oligo-anovulation without hyperandrogenism. Hyperandrogenic ovulatory women with PCO appear to have similar metabolic and reproductive features, albeit at a modest level, compared with classic PCOS patients with hyperandrogenic chronic anovulation. However, data are scarce regarding phenotypic features of normoandrogenic oligo-ovulatory women with PCO. The 2003 Rotterdam criteria will potentially increase the number of individuals identified as having PCOS, as it will capture the phenotypes that appear to represent the milder forms of the syndrome. Studies assessing the metabolic and reproductive characteristics of the different subgroups of PCOS patients diagnosed by Rotterdam criteria are urgently needed. It is of utmost importance that the applied diagnostic criteria in clinical studies of PCOS are thoroughly documented for the interpretation and comparability of the results.