Luteal phase sertraline treatment of premenstrual dysphoric disorder (PMDD): Effects on markers of hypothalamic pituitary adrenal (HPA) axis activation and inflammation

Rationale

Premenstrual dysphoric disorder (PMDD) is characterized by severe affective symptoms during the luteal phase of the menstrual cycle. There is some evidence of altered interactions between the hypothalamic pituitary gonadal (HPG) and hypothalamic pituitary adrenal (HPA) axes in PMDD. There is also evidence that similar affective disorders such as major depression and perinatal depression are associated with dysregulation in immune factors, but this has not been characterized in PMDD.

Aims

The goals of this exploratory study were to identify 1) whether HPA-HPG axis interactions and immune markers differ between PMDD patients and controls across the menstrual cycle; 2) how luteal phase sertraline treatment impacts stress and inflammatory markers.

Methods

Participants were females age 18–50 with regular menstrual cycles, not using psychotropic or hormonal medications, and were assigned to a control group or PMDD group based on prospective daily symptom ratings and clinical interview. Blood was drawn in the follicular and luteal phases, during laboratory sessions involving a mildly stressful task. In a second luteal phase, PMDD participants received open-label sertraline (50 mg/d) from ovulation to menses. Serum cortisol and ACTH were measured via ELISA and operationalized as area under the curve with respect to ground (AUCg), and peak level following laboratory task. Serum TNF-α, IL-6, CXCL-8, and IL-1β were measured using multiplex kits. Serum allopregnanolone (ALLO) was measured by gas chromatography/mass spectroscopy. To characterize HPA-HPG axis interactions across the menstrual cycle in PMDD participants and controls, multilevel linear models predicted cortisol and ACTH from the interaction of cycle phase (controlling for sertraline treatment), ALLO, and group. To determine the effects of sertraline treatment on inflammatory markers and how groups might differ in cyclical change on each marker, multilevel linear models predicted inflammatory markers from cycle phase (controlling for sertraline treatment) and group. A final set of exploratory models tested whether inflammatory markers predict premenstrual symptom score severity.

Results

The sample included n=77 participants (41 controls, 36 PMDD); 28 participants with PMDD completed sertraline treatment. Group x phase x ALLO interactions showed that higher ALLO levels predicted lower cortisol peak in the treated luteal phase (interaction between phase and ALLO, p=0.042), and there was a higher cortisol peak in the treated luteal phase than the untreated luteal phase (p=0.038). CXCL-8 was significantly associated with premenstrual symptom severity after controlling for group and cycle phase (p=0.011). There were no main effects of group, phase, or ALLO on cortisol AUCg, ACTH AUCg, IL-6, CXCL-8, IL-1β, nor TNF-α (p’s>0.05).

Conclusion

Serum markers of HPA axis and immune function did not vary by menstrual cycle phase nor PMDD status. However, sertraline treatment in the luteal phase was associated with higher ALLO levels predicting lower cortisol peak in response to mild laboratory stress, suggesting that sertraline treatment may normalize HPG-HPA axis interactions among individuals with PMDD. Greater premenstrual symptomatology was associated with higher levels of the inflammatory marker CXCL-8, but further research is needed into the potential role of inflammation in PMDD.