Background: The polycystic ovary syndrome (PCOS), a condition defined by hyperandrogenism and ovulatory dysfunction, is the most common endocrinopathy in women of reproductive age. Women classified as having PCOS according to the most recent consensus statement are heterogeneous both clinically and biochemically, and are therefore likely to have differing aetiologies, long term sequelae and possible treatment options.
Aim: To utilise principal components analysis in a cohort of women with PCOS, aiming to define orthogonal factors associated with the syndrome that may be of use delineating subgroups within PCOS.
Methods: A retrospective chart review studying data from a clinic population of 261 women with PCOS, as defined by the 1990 NIH criteria, mean age 29 (±8.8 yrs). SPSS 11.0 was used for analysis.
Results: Three factors that explained 61% of variance were recognised. (See table). Factor one includes the components reflective of insulin resistance and obesity, factor two includes hypertension and dyslipidaemia, and factor three includes higher LH:FSH ratio and higher plasma testosterone. Factor analysis results in PCOS*
|Factor 1||Factor 2||Factor 3|
|↑ HOMA-IR (0.64)||↑ Systolic BP (0.54)||↑ Testosterone (0.77)|
|↑ Waist circumference (0.80)||↑ LDL (0.71)||↑ LH:FSH ratio(0.63)|
|↑ Body mass index (0.79)||↑Triglycerides(0.72)|
|↓ HDL (−0.72)|
|↓ SHBG (−0.74)|
|*(Factor loadings >0.5 are reported)|
Conclusion: Three independent factors within a cohort of PCOS women have been identified using principal components analysis. The findings suggest different pathogenetic entities within this syndrome that may have both prognostic and therapeutic implications for the long-term health outcomes for women with PCOS.