Three hundred women with severe premenstrual syndrome will be divided into 3 groups using computer generated random numbers. Group 1 will receiveCombined oral contraceptives (COC) containing drospirenone (drospirenone 3mg+Ethinylestradiol 0.03mg) daily for 21 days starting from the 3rd day of menstruation in addition to oral fluoxetine 20 mg daily. Group 2 will receive fluoxetine 20 mg daily in addition to an oral placebo similar to COC daily for 21 days starting from the 3rd day of menstruation.Group 3 will receive oral placebo similar to COC daily for 21 days starting from the 3rd day of menstruation in addition to a daily oral placebo similar in size, color and structure to fluoxetine.

Premenstrual syndrome (PMS) manifests with distressing physical, behavioral and psychological symptoms, in the absence of organic or underlying psychiatric disease, which regularly recur during luteal phase of each menstrual cycle and disappear or significantly improve by the end of menstruation. Approximately 85-90 % of women may experience premenstrual emotional and physical changes in their reproductive age and the prevalence of severe PMS ranges from 3% to 8%.

The etiology of PMS is unknown but cyclical ovarian activity and the effect of estradiol and progesterone on serotonin and gamma-amino butyric acid are key factors. Absence of PMS before puberty, in pregnancy and after the menopause supports a role of cyclical ovarian activity in PMS etiology. PMS symptoms include psychological symptoms like mood swings, irritability, depression and feeling out of control; physical symptoms like breast tenderness, bloating and headaches; and behavioral symptoms like reduced visuospatial and cognitive ability. To diagnose PMS, symptoms should be recorded prospectively over two cycles using a symptom diary. Several symptom diaries exist but the Daily Record of Severity of Problems (DRSP) is reliable, simple for patients and recommended by the Royal College of Obstetricians and Gynecologists.

Life style modifications like stress reduction, exercise, and low carbohydrate diet have been suggested for management of PMS but have not been thoroughly investigated. Pharmacological treatments of PMS include COC, serotonin reuptake inhibitors gonadotrophin-releasing hormone agonists, anxiolytics, and diuretics. Various other treatments have also been recommended including acupuncture, dietary supplements, and bright light therapy, PMS treatment should be based on individual symptoms, concomitant medical history and the need of contraception.

The objective of the study is to investigate the role of the combined use of fluoxetine and COC versus fluoxetine in the treatment of PMS.

Three hundred women with severe premenstrual syndrome will be divided into 3 groups using computer generated random numbers. Group 1 will receive COC containing drospirenone (drospirenone 3mg+Ethinylestradiol 0.03mg) daily for 21 days starting from the 3rd day of menstruation in addition to oral fluoxetine 20 mg daily. Group 2 will receive fluoxetine 20 mg daily in addition to an oral placebo similar to COC daily for 21 days starting from the 3rd day of menstruation.Group 3 will receive oral placebo similar to COC daily for 21 days starting from the 3rd day of menstruation in addition to a daily oral placebo similar in size, color and structure to fluoxetine.

To diagnose PMS women will be asked to document their symptoms daily for 2 months using the daily record of severity of problems (DRSP). DRSP is a questionnaire comprised of 25 physical and emotional symptoms including impairment of physical and social activities, women will be asked to give a score of 1 to 6 for each symptom 1 = not at all, 2 = minimal, 3 = mild, 4 = moderate, 5 = severe, 6 = extreme. The investigators will add the symptoms scores of the first day of menses and PMS will be excluded if the score is< 50. If the total score is greater than 50, the investigators will recorded two cycles of symptoms. If more than three items have an average score of more than 3 (mild) during the luteal phase, the investigators will add the scores of five-day intervals during the luteal and follicular phases. PMS will be diagnosed when the luteal phase score is 30 percent greater than the follicular phase score in the 2 months. All women will have a psychiatric review and women with underlying psychiatric disease will be excluded. Women who are unable to interact socially or professionally because of the symptoms will be considered to have severe PMS. All married women will be advised to use barrier contraceptives.

All women will be asked to take the medications for 6 months and record their symptoms using the DRSP. Women with suboptimal social or professional interaction will be considered having moderate PMS. Women with no interference with the social or professional activity will be considered as having mild PMS. Women with day 1 DRSP score <50 and women with average luteal phase scores <30% more than the average proliferative scores will not be considered having PMS.

Inclusion Criteria:

• Severe PMS
• Regular periods

Exclusion Criteria:

• Underlying psychiatric disease.
• Body mass index > 35.
• Women on combined oral contraceptives.
• History of deep venous thrombosis, known liver, kidney or heart disease.
• Allergy to COC or fluoxetine