LC-ESI-MS/MS quantification of hair cortisol, cortisone and DHEA in combined oral contraceptive users

Abstract

Combined oral contraceptives (COCs) are used chronically by many women worldwide and in South Africa. Because many COC adverse effects coincide with symptoms of stress-related illnesses, including insulin resistance and hypertension, it is believed that these adverse effects are a result of chronic biological stress from long-term COC use and that COC users are at greater risk of developing stress-related illnesses. Different combinations of estrogens and progestins in COCs have varying effects on the stress hormone, cortisol, and stress-related symptoms. Previous studies often do not differentiate between COC compositions and most lack appropriate long-term measurement of biologically active free cortisol. In this study hair cortisol, as a measure of long-term free cortisol, was compared between users and non-users of the commonly used COC consisting of ethinyl estradiol (EE) and drospirenone (DRSP) to determine if COC use results in measurable chronic stress. To get a comprehensive reflection of total cortisol exposure, cortisone, the inactive form of cortisol was quantified as well. Dehydroepiandrosterone (DHEA), a hormone that has anti-cortisol activity, was also intended to be added to reflect total cortisol exposure. An in-house LC-ESI-MS/MS method was developed to simultaneously quantify cortisol, cortisone, and DHEA in hair. The use of ammonium fluoride as a mobile phase modifier enhanced the sensitivity of all analytes without derivatisation. This was, however, insufficient for the low levels of DHEA in hair. Derivatisation was then used to improve DHEA sensitivity but was unfavourable for cortisol. Since cortisol was the most important metabolite in this study, the non-derivatised method was used and DHEA had to be excluded. There were no statistical differences in hair cortisol, cortisone, total glucocorticoids, or the cortisol/cortisone ratio between COC users and non-user controls. These findings suggest that long-term use of COCs containing EE and DRSP does not result in chronic stress or increase the body’s exposure to cortisol through altered cortisol synthesis or overall cortisol activity. Further research is needed to understand the mechanisms behind stress-related COC adverse effects. As a secondary aim ethnicity was investigated as a covariate for chronic stress markers in hair. Only the hair cortisol/cortisone ratio differed between Black Africans and Caucasians (p = 0.05, Cohen’s d = 1.08), indicating that it should be adjusted for in future studies, particularly in the South African population.