Visual-spatial cognition in women with polycystic ovarian syndrome: the role of androgens.

STUDY QUESTION

Are women with polycystic ovary syndrome (PCOS) better at three-dimensional mental rotation than other women?

SUMMARY ANSWER

Women with PCOS scored significantly higher on a mental rotation task than a female control group.

WHAT IS KNOWN ALREADY

PCOS is a condition characterized by elevated testosterone levels. Some researches have found that three-dimensional mental rotation task performance is positively correlated with testosterone levels.

STUDY DESIGN, SIZE, DURATION: This cross-sectional study was conducted between June 2006 and January 2009. The participants were 69 women with PCOS and 41 controls recruited from five gynaecology clinics in London. The control group consisted of non-PCOS women of comparable subfertility to PCOS group. These groups sizes gave roughly 80% power to detect moderate effect sizes for the main statistical test.

PARTICIPANTS/MATERIALS, SETTING, METHODS: Participants were recruited at London gynaecology clinics. The women were aged between 18 and 43. PCOS was diagnosed based on the Rotterdam criteria. Controls were women who experienced some degree of subfertility. Blood samples from participants were frozen for up to 4 months until being assayed by direct electrochemiluminescence. The mental rotation task was undertaken electronically. Some questionnaires and other tasks were completed as control measures.

MAIN RESULTS AND THE ROLE OF CHANCE

Women with PCOS scored significantly higher than controls: median (range) 3.00 (0-9) and 2.00 (0-8), respectively (U = 1147.500, N1 = 69, N2 = 41, P < 0.047). Within the PCOS group, circulating levels of testosterone were significantly positively correlated with three-dimensional scoring (rs = 0.376, n = 56, P < 0.002), whereas estradiol was significantly negatively correlated with three-dimensional scoring (rs = -0.473, n = 29, P < 0.010). In the control group, the relationship between sex hormones and mental rotation was non-significant. Other factors, including general intelligence and social class, did not account for these findings. A subgroup analysis comparing hyperandrogenic PCOS cases, non-hyperandrogenic PCOS cases and controls, in which age and body mass index were controlled for using ANCOVA, found a non-significant difference in three-dimensional scoring between the three groups (F = 1.062, d.f. = 1, 73, P < 0.351).

LIMITATIONS, REASONS FOR CAUTION: The small number of women in the control group meant that correlations were underpowered in this group.

WIDER IMPLICATIONS OF THE FINDINGS

This study is the first to find a benefit of PCOS in visuospatial cognition, and the first to find a link between visuospatial cognition and sex hormones in PCOS. The fact that the correlations went in the opposite direction in the PCOS group compared with the controls might suggest the influence of increased prenatal exposure to androgen in PCOS.

STUDY FUNDING/COMPETING INTEREST(S): The assays for this study were funded by the Department of Psychology, City University London. All authors report no conflicts of interest.