Oral contraceptives and premenstrual symptoms: Comparison of a 21/7 and extended regimen

Transcription

1 American Journal of Obstetrics and Gynecology (2006) 195, Oral contraceptives and premenstrual symptoms: Comparison of a 21/7 and extended regimen Andrea L. Coffee, PharmD,* Thomas J. Kuehl, PhD, Sherilyn Willis, MD, Patricia J. Sulak, MD Department of Obstetrics and Gynecology, Scott & White Memorial Hospital, Texas A&M University System Health Science Center College of Medicine, Temple, TX Received for publication February 14, 2006; accepted May 4, 2006 KEY WORDS Extended regimen of oral contraceptives Premenstrual symptoms Drospirenone Objective: The purpose of this study was to assess the incidence and severity of premenstrual-type symptoms in patients converted from a 21/7 oral contraceptive (OC) regimen to an extended regimen. Study design: This was a single center prospective analysis of the single item Scott and White (S&W) Mood Scale and the Penn State Daily Symptom Report (DSR17) during a 21/7-day followed by a 168- day extended regimen of an OC containing 3 mg of drosperinone and 30 mg of ethinyl estradiol (DRSP/EE). Results: Of the 114 patients who began the study, 111 completed the preextension 21/7 phase of the study. There were significant differences in severity in the DSR17 and the S&W mood scale among days of the cycle. (P!.0001) The highest values in both scales occurred during the 7-day hormone free interval (HFI) of the 21/7 cycles (P!.001). Of the 111 patients who completed the 21/7 phase of the study, 102 (92%) completed the 168-day extended regimen. During the extended phase of the study, subjects were divided into 2 groups: those with a 100% increase in symptoms from the first half to the second half of the last 21/7 cycle were labeled as high cyclic variability, whereas those with lesser or no cyclic change were labeled as low cyclic variability. There were 55 (54%) with increased cyclic variability in mood scores peaking during the 7-day HFI. Premenstrual-type symptoms measured by both the S&W mood scale and the DSR17 instrument decreased during the extended DRSP/EE OC regimen (P!.0001) compared with the preceding 21/7 cycle, with the greatest improvement detected in the sixth month of continuous OCs (P!.003). The patient group with greatest cyclic variability during the 21/7 regimen demonstrated the most improvement during the 168-day regimen (P!.0001). The single item S&W mood scale was significantly (P!.05) correlated to each of 17 elements of the DSR17 with Spearman R correlation coefficients of 0.25 to The greatest correlation coefficient (Spearman s R = 0.66) is with the sum of all 17 items. Conclusion: A 168-day extended regimen of DRSP/EE led to a decrease in premenstrual-type symptoms compared with the 21/7-day regimen. Ó 2006 Mosby, Inc. All rights reserved. Funding provided by Berlex Laboratories, Montville, NJ. * Reprint requests: Dr Andrea L. Coffee, Department of Obstetrics and Gynecology, Scott & White Clinic, 2401 South 31st Street, Temple, TX acoffee@swmail.sw.org Premenstrual symptomatology is common in reproductive aged women and can greatly influence quality of life. Ovulation is a prerequisite of premenstrual disorders with symptoms only occurring during the luteal phase of ovulatory cycles then disappearing after the onset of /$ - see front matter Ó 2006 Mosby, Inc. All rights reserved. doi: /j.ajog

2 1312 Coffee et al menstruation. Mild symptoms, which usually do not interfere with daily activities, are experienced occasionally by almost all ovulatory women and include breast tenderness, bloating/swelling, and pelvic heaviness. Premenstrual syndrome (PMS) is a premenstrual disorder characterized by bothersome adverse somatic and/or affective symptoms. 1 Premenstrual dysphoric disorder (PMDD), as defined by the American Psychiatric Association, is the most severe type of premenstrual disorder associated with significant impairment. 2 Studies have documented the detriment in quality of life parameters and the economic toll in lost productivity and treatments. 3,4 Several studies have assessed the effects of standard 21 day/7 day oral contraceptive (OC) regimens on premenstrual symptoms with inconsistent results, often showing no or minimal improvement. 5-7 The standard 21/7 design may actually induce symptoms. Menstrual-related symptoms including headache, mood swings, abdominal cramping, bloating, and breast tenderness are long recognized side effects associated with OCs. Studies have documented an increase in premenstrual symptoms during the last week of active pills extending into the 7-day hormone free interval (HFI). 6,8 Lack of ovarian suppression during the standard 7-day HFI leads to endogenous estradiol production. Endogenous estradiol levels have been documented to rise at the end of the 7-day HFI, peak in the first half of the active pill cycle, and then decline during the last week of active pills. 9,10 This decline in endogenous estradiol levels during the last week of active pills may be responsible for the estrogen-withdrawal symptoms beginning to appear during the last week of active pills with further exacerbation of premenstrual type symptoms during the subsequent 7-day HFI. 6,8 A novel OC with the progestin drosperinone has been shown in clinical trials to significantly improve symptoms of PMS and PMDD apparently because of its unique antimineralocorticoid activity As opposed to all other progestins in today s OCs, which are derivatives of 19-nortestosterone, drosperinone is derived from spironolactone. 14 Spironolactone is the only diuretic shown in randomized placebo controlled trials to improve the symptoms of PMS. 15,16 While studies of extended OC regimens compared with standard 21/7 day regimens have primarily assessed bleeding patterns, a review of the literature on extended OCs failed to identify any such prospective studies comparing PMS type symptoms with a validated instrument. 17 The present study was conducted to assess the exact timing and severity of premenstrual type symptoms in OC users during standard 21/7 regimens followed by an extended OC regimen using 2 daily symptomatology instruments. Material and methods This single center, prospective cohort study was approved by the Scott & White Institutional Review Board and written informed consent was obtained. At study entry, all subjects were current combination OC users of at least 3 months duration, between the ages of 18 and 48. They had no contraindications to continuing OCs, including no history of myocardial infarction, stoke, uncontrolled hypertension, venous thromboembolic events, breast cancer, diabetes mellitus, and liver disease. Advertisements explained the study concept of suppression of monthly bleeding and associated symptoms with an extended OC regimen and were placed throughout the hospital, outlying clinics, and in local newspapers. Subjects were recruited from September 2003 to April 2004 into an extended OC study of 168 hormone containing pills (24 consecutive weeks) of a monophasic OC containing 3 mg of drosperinone and 30 mg of ethinyl estradiol (DRSP/EE) in each active tablet (Yasmin, Berlex Labs, Montville, NJ). Before the 168 day extended phase, patients completed 2 21/7-day regimens of DRSP/ EE. Patients entering the study on an OC containing any of the 19-nortestosterone progestins completed 1 month of that OC before the 2 21/7 DRSP/EE cycles. During the 21/7 cycles and the extended cycle, subjects were asked to complete 2 symtomatology scoring instruments daily: a single item mood scale and a validated comprehensive multi-item scale. The Scott and White (S&W) Mood Scale was piloted in this study to evaluate its effectiveness as a simple single item assessment of mood lability representing a composite of anxiety, depression, and irritability scored on a scale of 0 to 10. A score of 0 denoted none of these symptoms present while a score of 10 represented worst emotional symptoms ever experienced. Subjects also completed the Penn Daily Symptom Report (DSR17), consisting of 17 physical or emotional symptoms. 18 The severity of each element is rated on a scale of 0 to 4: Zero is defined as not present at all; 1 as mild, only slightly apparent; 2 as moderate, aware of symptoms, but doesn t affect daily activity at all; 3 as severe, continuously bothered by symptoms; and 4 as very severe, symptom is overwhelming and/or interferes with daily activity. Elements rating physical components or symptoms include fatigue, poor coordination, headaches, aches, swelling/ bloating/weight gain, cramps, and breast tenderness. Elements rating behavioral components include feeling overwhelmed, feeling hopeless, anxiety, irritability, mood swings, food cravings, decreased interest in usual activities, depression, insomnia/hypersomnia, and difficulty concentrating. After completing the 21/7 cycles, patients began the 168 days of continuous active pills of DRSP/EE. The only exception allowed to the 168 consecutive continuous active pills were patients who experienced breakthrough spotting or bleeding at least 7 consecutive days who were randomized to continuing active pills or taking a 3 day HFI. Details of bleeding data have been published. 19

3 Coffee et al 1313 Figure 1 Comparison of daily scores for 17 items using DSR17 in patients initially on a 19-nortestosterone (19-NorT) progestin OC switching to DRSP/EE for cycles 2 and 3. Cycles and days within cycles differed (P!.0001). However, cycle 2 did not differ (P =.14) from cycle 3 for these patients. Statistical analysis Demographic variables including age, height, and weight were expressed as means with standard deviations. Because of previously published studies with 21/7 regimens documenting an increase in premenstrual symptomatology during the last week of active pills peaking in severity during the 7 day HFI, 6,8 patients were divided into 2 groups to assess whether those reporting increased severity of symptoms during the second half of the 21/7 regimen (last 7 active pills plus 7 day HFI) had a different response to the extended regimen compared with those who reported minimal to no cyclic variation in symptoms. Values for 2 groups based on variability in mood rating from the first 2 weeks to the second 2 weeks were identified during the last 21/7 DRSP/EE OC regimen and labeled as either low cyclic variability or high cyclic variability in mood score. Patients who had a 100% or greater increase in the average mood score from the first 2 weeks to the second 2 weeks of the last 21/7 OC cycle were labeled as having high cyclic variability. Demographic variables were examined using analysis of variance. Homogeneity of variances for quantitative variables was examined and, when needed, data compared with nonparametric tests including Friedman analysis of variance, Wilcoxon matched pairs test, and Mann-Whitney U nonparametric test. To compare the last 21/7 cycle with the 168-day cycle, the extended regimen was divided into 6 28-day periods equaling 168 days. Results Of the 114 patients who entered the study, 111 completed the 21/7-phase study and 102 completed the extended phase requirements including both daily scoring instruments. Fifty-four of 102 subjects completing the study were on a 19-nortestosterone containing progestin OC when they entered the study while the remainder were on DRSP/EE at study initiation. Patients entering the

4 1314 Coffee et al Table I Characteristic Enrollment characteristics and subgroup* analysis during the 21/7 cycle Total group, mean (SD) and range Low cyclic variability, mean High cyclic variability, mean P-value y n Age in years 31.9 (7.0) 18.0 to Height in inches 64.7 (2.7) 58.5 to Weight in pounds 157 (32) 105 to BMI in kg/m (4.7) 17.2 to Using medication 17/102 (17)% 7/47 (15%) 10/55 (18%).66 for anxiety and/or depression Average daily S&W score 1.3 (1.4) 0 to Average S&W scores of the first 2 weeks 0.98 (1.4) 0 to Average S&W scores of the last 2 weeks 1.68 (1.6) 0 to Average daily sum of DSR (6.3) 0 to Average DSR scores of the first 2 weeks 4.1 (6.0) 0 to Average DSR scores of last 2 weeks 7.6 (7.8) 0 to * Subgroups defined by a 2-fold change in Scott & White (S&W) Mood Scale between first 2 weeks and second 2 weeks of last 21/7 cycle or an average daily increase of more than 0.2 units in daily scale of 0 to 10 for patients with a 0.0 score during the first 2 weeks. y Subgroup comparison for normally distributed demographic variables performed with multivariate analysis of variance; for proportions with Pearson s chi-square for proportions; and S&W Mood Scale and 17-item DSR scale with Mann-Whitney U nonparametric test. study on DRSP/EE OCs did not statistically differ (P =.34) in DSR 17 scores of their initial cycle (6.8 G 0.9 for mean G SE) compared with patients entering on 19-nortestosterone progestin OCs (8.1 G 0.9). Caution must be taken in comparing the DSR 17 scores of the various progestin pill types at study entry, as this was not a randomized comparison of these OC types. Also, length of previous OC use was not recorded but in general, patients entering on a 19-nortestosterone containing OC were longer term users than those entering on the newer DRSP/EE OC. Of the 54 subjects entering the study on a 19-nortestosterone progestin OC and completing the entire study requirements, the average DSR17 scores did not differ (P =.59) based on the type of 19-nortestosterone progestin OC (norethindrone, levonorgestrel, desogestrel, norgestimate), although the numbers of each pill type were small. Patients switching from a 19-nortestosterone progestin OC to DRSP/EE did demonstrate a decrease (P!.0002) in symptoms (Figure 1) during the 2 subsequent cycles of DRSP/EE. However, the pattern of symptom scores continued during the DRSP/EE cycles with the highest DSR17 scores during the HFI days 23 through 27 (P!.001). Of these 111 who began the extended cycle, 102 (92%) completed the 168-day extended DRSP/EE cycle regimen. Details of the 9 subjects who withdraw from the extended phase have been previously reported, with most because of study compliance issues and breakthrough bleeding. 19 Characteristics of the 102 subjects completing all study requirements of the 168-day extension phase are shown in Table I. The subgroup of patients with a 100% increase in cyclic variations in mood score from the first half to the second half of the last 21/7 cycle included 55 (54%) of those completing the trial, while 47 (46%) did not meet the set criteria for cyclic variation. The 2 subgroups did not differ in age, weight, height, or body mass index. Seventeen of the 102 patients reported using medications for anxiety and/or depression during 21/7 OC cycles. There was no difference in the proportion of patients using these medications among patients within the 2 subgroups (Table I). Subgroups also did not differ in their average daily scores for either the single item S&W mood scale or the DSR17 during the entire 21/7 cycle. However, the subgroups did differ in both of these scales for both the first 2 weeks and the last 2 weeks of the 21/7 cycle (Table I). Scores also varied within subgroups between the first 2 weeks and the last 2 weeks of the 21/7 cycle. For example, the high cyclic variability subgroup had an average DSR17 score of 2.2 and a S&W mood score of 0.4 during the first 2 weeks of the 21/7 cycle that differed (P!.0001) using Wilcoxon matched pairs test from 8.2 and 2.2, respectively, during the last 2 weeks of the 21/7 cycle. However, the low cyclic variability subgroup had an average DSR17 score of 6.3 for the first 2 weeks that did not differ (P =.31) from that of 7.0 for the second 2 weeks. This subgroup had an average S&W mood score of 1.6, which was actually slightly greater (P =.012) than the score of 1.4 in the second 2 weeks. Figure 2 demonstrates the difference in response patterns of the 2 subgroups during the transition from the 21/7 DRSP/EE OC cycles to a 168-day DRSP/EE extended regimen. Symptom ratings increased during the third week of each standard 21/7 regimen in both groups. Patients in the high cyclic variability group have significantly (P =.016) greater symptom scores during the HFI (10.4 G 0.4 vs 9.0 G 0.4, mean G SE) than patients in the low cycle variation group. During the extension phase, those with high cyclic variability had lower (P!.0001) symptom scores compared with

5 Coffee et al 1315 Figure 2 Comparison of daily scores for 17 items using the DSR17 for the 2 21/7 DRSP/EE OC cycles followed by the 168 day extended DRSP/EE OC regimen for 55 patients with high cyclic variations in mood during last 21/7 cycle and 47 with low cyclic variations. Groups differ (P!.0001), treatment days differ (P!.0001), and group by treatment day interactions are significant (P =.042). those with less cyclic variation (3.2 G 0.1 vs 4.9 G 0.1). During the third week of the extended regimen, both groups demonstrate a similar rise in symptoms to peak levels on day 19 of 6.4 G 0.9 for high variability group and on day 20 of 7.0 G 1.0 for the low variability groups followed by decreases to similar levels (P =.33) during the fourth week of continuous OCs to 4.6 G 0.4 and 5.2 G 0.3 for high and low variability subgroups, respectively. Figure 3 highlights the changes associated with the 28 days of the last DRSP/EE 21/7 OC cycle and the first 28 days of an extended cycle for all 102 patients. No difference (P O.05) is seen until day 22, when patients in the 21/7 OC cycle have entered the 7-day HFI and patients in the extended OC regimen have continued active OCs. Table II compares the single item S&W mood score and the DSR17 rating for the last 21/7 OC cycle and each of 6 28-day segments of the 168-day extended regimen. Premenstrual-type symptoms decrease (P!.0001) in all segments of the extended regimen compared with the standard cycle. Also, the fifth and sixth segments demonstrate a further decrease from the first segment, suggesting that increased duration of an extended regimen of DRSP/EE OC can further reduce premenstrual symptoms, especially in those patients with cyclic variation (Figure 2). Table III compares each of the 17 items of the DSR scale during the 21/7 cycle and the last 28 days of the extended cycle. Differences in premenstrual-type symptoms occur in 16 of the 17 items rated by the DSR17, with decreased interest the only parameter evaluated that was not significantly altered by the extended regimen. The single item S&W Mood Scale (0-10) is significantly (P!.05) correlated to all 17 elements of the DSR17 with Spearman R correlation coefficients of 0.25 to Correlations are highest with 3 items

6 1316 Coffee et al Figure 3 Comparison of 28 days of the last 21/7 DRSP/EE OC regimen and the first 28 days of the extended DRSP/EE OC regimen demonstrating similar variation in daily pattern of the DSR17 for the first 21 days. However, beginning on day 22, the patterns significantly (P!.05) diverge as symptoms continue to increase during the hormone free interval of the 21/7 regimen but decrease with continued active OCs in the extended regimen. (anxiety, irritability, and mood swings with Spearman R correlation coefficients of 0.50, 0.50, and 0.57, respectively) using the data set from the standard 21/7-day OC cycle before the 168-day extended regimen. The largest correlation coefficient (Spearman s R = 0.66) is with the sum of all 17 items of the DSR17. Comment The present study confirms that many women experience a multitude of premenstrual type symptoms while utilizing standard 21/7 OCs. Regardless of the progestin content of the pill, symptoms worsen during the last week of active pills, peaking in intensity during the 7-day HFI. The occurrence and severity of symptoms interestingly coincides with the decline in endogenous estradiol levels during the last week of active pills into the beginning of the HFI. 9,10 While occurrence and severity of symptoms during the 21/7 cycles was demonstrated to increase during the last week of active pills and further increase during the HFI with all types of progestin containing OCs, a statistically significant decrease in DSR17 ratings was noted when switching from a 19-nortestosterone progestin OC to a DRSP OC. This could be secondary to the antimineralcorticoid effects of DRSP, or could be simply a placebo effect as this trial was not controlled for order of pill use during the 21/7 cycle. Elimination of the 7-day HFI with the extended 168-day DRSP/EE OC regimen resulted in a statistically significant reduction (P!.001) in premenstrual type symptoms. Satisfaction with the DRSP/EE regimen was high as attested to by the high completion rate (92%) of the extended regimen. The statistically significant reduction in symptoms was noted to be more pronounced in the last 2 28-day intervals of the 168-day extended regimen. While a randomized trial would be

7 Coffee et al 1317 Table II Comparison of S&W Mood Scale and sum of DSR17 item symptom scale during the 21/7 OC cycle and 6 28 day intervals of the 168 day extended OC regimen (n = 102) Average rating of S&W Mood Scale of 0 to 10 Average rating of sum of DSR17 scale of 0 to 68 OC study interval Median (mean G SD) P value* Median (mean G SD) P value y 21/7 OC cycle 0.93 (1.3 G 1.4) 3.8 (6.0 G 6.4) Six 28 day intervals of 168-day extended OC regimen First 0.46 (0.97 G 1.3)! (4.2 G 4.8)!.0001 Second 0.43 (0.86 G 1.4)! (4.0 G 5.4)!.0001 Third 0.36 (0.86 G 1.4)! (4.2 G 6.0)!.0001 Fourth 0.32 (0.86 G 1.4)! (4.1 G 5.7)!.0001 Fifth 0.21 (0.74 G 1.3)! (3.5 G 5.0)!.0001 Sixth 0.17 (0.67 G 1.2)! (3.3 G 4.7)!.0001 * Intervals within patients compared to 21/7 OC cycle using Friedman s analysis of variance for nonparametric variables. The six 28 day intervals of the 168-day extended OC regimen differ (p! ) from the 21/7 OC cycle and differ between intervals (p! ). The fifth and sixth intervals differ from the first interval (p! 0.001) using Wilcoxon matched pairs test. y The 6 intervals of extended OC regimen differ (P!.0001) from 21/7 OC cycle and between intervals (P!.0001). The fifth and sixth intervals differ from the first interval (P!.003). Table III Comparison of averages for each item of the DSR17 during the 21/7 cycle and the last 28 day interval of the 168-day extended regimen Characteristic of individual items Standard 21/7 OC cycle Sixth 28-day interval of extended regimen P value* Fatigue 0.65 G G 0.05!.0001 Poor coordination 0.19 G G 0.02!.0001 Feeling overwhelmed 0.31 G G 0.03!.0001 Feeling hopeless 0.19 G G Headache 0.53 G G 0.04!.0001 Anxiety 0.43 G G 0.04!.0001 Aches 0.43 G G 0.04!.0001 Irritability 0.39 G G 0.04!.0001 Moods swings 0.46 G G 0.04!.0001 Swelling, bloating, weight gain 0.47 G G 0.06!.0001 Food cravings 0.39 G G Decreased interest in usual activities 0.20 G G Cramps 0.18 G G Depression 0.35 G G Breast tenderness 0.25 G G Insomnia/hypersomnia 0.31 G G Difficulty concentrating 0.25 G G * Comparison of 2 measures within patients for each item of DSR17 using Wilcoxon matched pairs tests. necessary to determine if 1 progestin type OC given in an extended fashion has a greater benefit in reducing PMS type symptoms, the unique antimineralcorticoid properties of DRSP have been shown to provide greater reduction in severity of PMS/PMDD compared to other type progestin OCs. 12,13 Premenstrual symptomatology was assessed in this study by 2 instruments: the single item S&W Mood Scale and the multi-item Penn State DSR17. Detailed daily multiresponse questionnaires can be time consuming and cumbersome. Our single score S&W Mood Scale of a composite of irritability, depression, and anxiety was found to be comparable to the comprehensive 17 item DSR scale. If general mood is the main parameter to be evaluated in a study or in clinical practice, this simple scale is easy for patients to monitor daily and may be comparable to more time-intensive multiquestion instruments that are difficult to implement in a busy clinical practice and in hectic patient s lives. Other studies are needed to further assess its reliability as a validated mood scale instrument. Two unique groups of patients were identified and evaluated within the study cohort with distinctly different scores during the 21/7 and the extended regimen. A 100% increase in scores during the first half compared with the second half of the 21/7 cycle was arbitrarily set

8 1318 Coffee et al to identify a population with highest cyclic variability. The group with high cyclic variability differed from the low cyclic variability group not only with regard to change in scores from the first to the second half of the 21/7 cycle (as set by the study definition), but also in comparing the scores in the first half of the 21/7 cycle. Subjects in the low cyclic variability group had significantly (P!.011) higher scores in the first half of the 21/7 cycle compared with the high cyclic variability group. By dividing the 102 patients into groups based on a 100% increase in cyclic variability between the first and second half of the last 21/7 cycle, a group resembling premenstrual syndrome was identified. In this high cyclic variability group, symptoms are directly correlated with the known fluctuations seen with the 21/7 cycle and are ameliorated to a greater degree in this group with elimination of the hormonal fluctuations with the extended cycle. Subjects with low cyclic variability during the 21/7 cycle did show improvement with the extended regimen but to a lesser degree, possibly because of at least 3 diverse groups being represented in this subgroup: (1) cyclic variation from first half to second half of the cycle but less than the defined 100% increase; (2) no cyclic variation with low scores throughout the cycle; or (3) no cyclic variation but high scores throughout the cycle. Those with cycle variation of a lesser degree could be responsible for the improvement seen during the extended cycle in the low variability subgroup. If we define response to the extended DRSP/EE OC regimen as a 50% decrease in average DSR17 scores for the last 28 day interval of the 168 day extended regimen compared with the last 21/7 DRSP/EE cycle, 33 (60%) of the high variability group and 25 (53%) of the low variability group would be responders. (P =.49 Pearson s chi-square). Our study was designed to only assess the response of an extended regimen in those patients with or without a 100% increase in severity of symptoms during the 21/7 cycle. A larger trial would be necessary to assess the best treatment options for multiple subgroups of patients with various degrees of symptomatology throughout the cycle with and without cyclic variation. Those with no cyclic variation but elevated scores throughout the 21/7 cycle may represent some patients with characteristic endogenous type depression where the multiple mood items evaluated are not influenced greatly by fluctuations in hormone levels or by the removal of these fluctuations with an extended regimen. This would be consistent with our knowledge of the biologic differentiation of endogenous depression compared with PMS/PMDD. If a patient truly has PMS type symptoms that primarily occur in the second half of the 21/7 cycle, then eliminating the 7-day HFI can significantly improve these often bothersome symptoms that can greatly affect quality of life. On the other hand, if significant mood symptoms are present throughout the 21/7 cycle with little to no increase in severity during the second half of the cycle, extended OC regimens may offer less benefit. These patients with significant symptomatology throughout the 21/7 cycle interfering with quality of life may respond better to antidepressants. Since many patients discontinue OCs because of premenstrual type symptoms including breast tenderness, bloating, headaches, and mood swings, warning patients of their possible occurrence and management options can be crucial in OC continuation. Our study along with others confirms that 21/7 regimens can induce PMS type symptoms. Extended regimens may reduce these problems. Knowing this, 2 issues need to be addressed. First, patients currently on 21/7 OCs need to be questioned regarding symptoms occurring during the end of the active pills into the HFI. Our experience with a simple 0-10 scale for rating mood on a daily basis during 1 or more 21/7 cycles can assist with this documentation in a clinical setting where detailed daily symptom logs are often not practical. Those with significant cyclic variation in symptomatology can be offered the option of an extended regimen. Patients initiating a 21/7 regimen also need to be informed of the documented increase in symptoms associated with the HFI and instructed to report their occurrence. Second, these results provide further evidence of the need to redesign OCs with modification of the standard monthly 7-day HFI. Data continue to accumulate on the lack of ovarian suppression and resultant follicular development, endogenous hormone production, withdrawal symptoms, and even ovulation with today s low dose OCs. 6,8,9,10 Fortunately, modifications in the basic 21/7-day design are being incorporated with extended cycle active pills, 20,21 added estrogen, 22 or a shortened HFI of less than 7 days. 9,23 A low dose OC with 24 active pills of ethinyl estradiol/drosperinone and a 4-day HFI has been shown to reduce the symptoms of PMDD. 11 Studies are needed to evaluate the effectiveness of extended OC regimens in women with documented PMS/PMDD. References 1. American College of Obstetricians and Gynecologists. Clinical management guidelines for obstetricians-gynecologists: premenstrual syndrome. ACOG practice bulletin 15. Washington, DC: ACOG; Premenstrual dysphoric disorder. In: American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision. Washington, DC: American Psychiatric Association; p Halbreich U, Borenstein J, Pearlstein T, Kahn LS. The prevalence, impairment, impact, and burden of premenstrual dysphoric disorder (PMS/PMDD). Psychoneuroendocrinology 2003;28(Suppl 3): Sternfeld B, Swindle R, Chawla A, Long S, Kennedy S. Severity of premenstrual symptoms in a health maintenance organization population. Obstet Gynecol 2002;99:

9 Coffee et al Graham CA, Serwin BB. A prospective treatment study of premenstrual symptoms using a triphasic oral contraceptive. J Psychosom Res 1992;36: Bäckstro m T, Hansson-Malmström Y, Lindhe BÅ, Cavalli-Bjorkman B, Nordenstrom S, et al. Oral contraceptives in premenstrual syndrome: a randomized comparison of triphasic and monophasic preparations. Contraception 1992;46: Joffe H, Cohen LS, Harlow BL. Impact of oral contraceptive pill use on premenstrual mood: predictors of improvement and deterioration. Am J Obstet Gynecol 2003;189: Sulak PJ, Scow RD, Preece C, Riggs MW, Kuehl TJ. Hormone withdrawal symptoms in oral contraceptive users. Obstet Gynecol 2000;95: Sullivan H, Furniss H, Spona J, Elstein M. Effect of 21-day and 24-day oral contraceptive regimens containing gestodene (60 mg) and ethinyl estradiol (15 mg) on ovarian activity. Fertil Steril 1999;72: Schlaff WD, Lynch AM, Hughes HD, Cedars MI, Smith DL. Manipulation of the pill-free interval in oral contraceptive users: the effects on follicular suppression. Am J Obstet Gynecol 2004; 190: Yonkers KA, Brown C, Pearlstein TB, Foegh M, Sampson-Landers C, Rapkin A. Efficacy of a new low-dose oral contraceptive with drospirenone in premenstrual dysphoric disorder. Obstet Gynecol 2005;106: Foidart JM, Wuttke W, Bouw GM, Gerlinger C, Heithecker R. A comparative investigative investigation of contraceptive reliability, cycle control and tolerance of two monophasic oral contraceptives containing either drospirenone or desogestrel. Eur J Contracept Reprod Health Care 2000;5: Sangthawan M, Taneepanichskul S. A comparative study of monophasic oral contraceptives containing either drospirenone 3 mg or levonorgestrel 150 mug on premenstrual symptoms. Contraception 2005;71: Krattenmacher R. Drospirenone: pharmacology and pharmacokinetics of a unique progestogen. Contraception 2000;62: Vellacott ID, Shroff NE, Pearce MY, Stratford ME, Akbar FA. A double-blind, placebo-controlled evaluation of spironolactone in the premenstrual syndrome. Curr Med Res Opin 1987;10: Wang M, Hammarback S, Lindhe BA, Backstrom T. Treatment of premenstrual syndrome by spironolactone: a double-blind, placebocontrolled study. Acta Obstet Gynecol Scand 1995;74: Edelman AB, Gallo MF, Jensen JT, Nichols MD, Schulz KF, Grimes DA. Continuous or extended cycle vs cyclic use of combined oral contraceptives for contraception. The Cochrane Database of Systematic Reviews 2005;Issue 3. Art. No: CD pub2. doi: / cd pub Batra P, Harper DM. Recognizing and treating premenstrual dysphoric disorder. JCOM 2002;9: Sulak PJ, Kuehl TJ, Coffee A, Willis S. Prospective analysis of occurrence and management of breakthrough bleeding during an extended oral contraceptive regimen. Am J Obstet Gynecol 2006 [Epub ahead of print]. 20. Anderson FD, Hait H. A multicenter, randomized study of an extended cycle oral contraceptive. Contraception 2003;68: Sillem M, Schneidereit R, Heithecker R, Mueck AO. Use of an oral contraceptive containing drospirenone in an extended regimen. Eur J Contracept Reprod Health Care 2003;8: Killick SR, Fitzgerald C, Davis A. Ovarian activity in women taking an oral contraceptive containing 20 microg ethinyl estradiol and 150 microg desogestrel: effects of low estrogen doses during the hormone-free interval. Am J Obstet Gynecol 1998;179: S Spona J, Elstein M, Feichtinger W, Sullivan H, Ludicke F, Muller U, et al. Shorter pill-free interval in combined oral contraceptives decreases follicular development. Contraception 1996;54:71-7.