Predictors of abnormal glucose metabolism in women with polycystic ovary syndrome.

OBJECTIVE

Polycystic ovary syndrome (PCOS) is a risk factor for type 2 diabetes mellitus and screening for abnormal glucose metabolism has been recommended by an oral glucose tolerance test (OGTT). This procedure is time-consuming and inconvenient, limiting its general use. Therefore, an easy method is wanted to separate PCOS women with normal from those with potentially abnormal glucose metabolism.

DESIGN

Simple parameters obtained from 101 consecutive PCOS patients were assessed by receiver operating curve analysis for their ability to predict abnormal glucose metabolism.

RESULTS

Comparing discriminating parameters at defined sensitivities revealed that, assessed by homeostasis model assessment (HOMA), insulin resistance (HOMA%S) had the highest specificity. At a cut-off point of 73.1%, HOMA%S had a sensitivity of 95.5% and a specificity of 51.9%. Applying this cut-off separated 59 women who had a high probability of abnormal glucose metabolism from 42 women who were at low risk (less than 2.5%). Fasting insulin was the second-best parameter and had a similar specificity. A screening strategy which applies HOMA%S or fasting insulin could almost halve the number of OGTTs by directing them to those PCOS women most likely to be suffering from abnormal glucose metabolism. The negative predictive value of this strategy was 97%. The strategy was tested and confirmed in a second and independent cohort of 264 PCOS women.

CONCLUSIONS

HOMA%S, or to a lesser extent fasting insulin, appears to allow for stratified metabolic screening of PCOS women with OGTT.