Characteristics of menstruation-related problems for adolescents and premarital women in Korea

Results: For primary amenorrhea, hypogonadotropic hypogonadism was more frequent in group I ( p = 0.007), and eugonadism in group II ( p = 0.0025). Chromosomal competent ovarian failure ( p = 0.003) and hyperprolactinemia ( p < 0.001) were more frequent causes of secondary amenorrhea in group II. Endometriosis without ovarian endometrioma was the more frequent laparoscopic finding for patients of group I ( p = 0.0429). Regarding AUB, dysfunctional uterine bleeding (DUB) was more frequent for group I ( p < 0.001) and endocrinopathies ( p = 0.006) and benign lesions of genital tract ( p < 0.0001) for group II. Conclusion: The menstruation-related problems showed different features for each group. These data might give us an insight, at least in part, into the menstruation-related problems of Korean young women. Keywords Adolescence Premarital women Menstruation-related problems 1 Introduction Menarche is a milestone for female puberty that signifies the approaching cessation of physical growth and the maturation of reproductive potential. This transitional era-like trait makes the disease features of adolescents very different from those of other age groups in the clinical field of obstetrics and gynecology. Moreover, the definition of normal menstrual function and the physician's approach for treating menstrual dysfunction may differ between the adolescent patients and the adult patients. For example, during the first several years of menarche, about 80% of menstrual cycles are irregular with a predictable range of 21–45 days duration; there are 2–7 days of bleeding and 3–6 pads are used daily [1] . Moreover, problems related with menstruation affect 75% of females by late adolescence and they are leading reasons for both school absenteeism [2] and visits to physicians [3] . Therefore, the American College of Obstetricians and Gynecologists recommends that young women schedule their first visits to obstetrician-gynecologists between the ages of 13 and 15 years. Although there are several studies reported on from North America or Europe [4–7] , adolescent gynecology remains a newly developing field in Korea. Moreover, according to a pediatric research undertaken in 1997 [8] , puberty begins by the age of 7 years in 3% of Caucasian girls and 6% of black girls. This supports the need for an adjustment of age for defining amenorrhea or precocious puberty according to the race and ethnicity of the patient. As a practical matter, it is not easy to obtain systemic data on menstruation-related diseases of Korean adolescents and premarital women because these patients in Korea tend to avoid visits to the gynecologists due to cultural reasons. Therefore, there has been persistent failure in the medical community to understand the general patterns of menstruation-related diseases. This study was undertaken to obtain insight into the characteristics of menstruation-related problems during adolescence, and we sought to evaluate the differences of these problems between adolescents and premarital women in Korea. 2 Materials and methods 2.1 Patients This study includes 1280 adolescent and premarital women who visited or were referred to the Young Lady Clinic (YLC) with menstruation-related problems at Samsung Medical Center (SMC) from February 1995 to July 2003. The patients were classified into two groups based on their age. The first group was comprised of adolescent patients aged 10–20 years, and the second group was comprised of premarital women aged 21–30. 2.2 Diagnostic criteria of menstruation-related disease Menstruation-related problems were categorized as the following: amenorrhea, oligomenorrhea, dysmenorrhea, abnormal uterine bleeding (AUB), and premenstrual syndrome (PMS). Primary amenorrhea was defined as not having menstruation until 14 years of age with the absence of secondary sexual characteristics or no menarche until 16 years of age. Secondary amenorrhea was defined as not having menstruation for more than 3 cycles or 6 months after menarche. Oligomenorrhea was defined as having infrequent menstrual cycles, and each cycle is longer than 35 days. Dysmenorrhea was classified as primary and secondary dysmenorrhea by the clinical characteristics, USG and such laboratory tests as the CA 125 levels. AUB was defined as any type of bleeding that was irregular in amount, duration or frequency. PMS was diagnosed on the basis of repeated cyclic physical and mental symptoms. Formal psychiatric consultation was undertaken to exclude the presence of psychiatric problems. 2.3 Evaluation of patients For the initial evaluation of amenorrhea and oligomenorrhea, we took a detailed history of the patients’ growth and development, menstruation and any associated medical problems. Complete physical examinations were conducted including the current height, weight and pubertal stage, and evaluation was undertaken for the signs of hyperandrogenism such as hirsutism, clitoromegaly and acne. For the differential diagnosis, we performed hormone studies such as gonadotropin (LH, FSH), sex hormones (estradiol, progesterone, free-testosterone), prolactin, β-hCG, thyroid stimulating hormone (TSH) and imaging studies such as the sella coned-down view, brain MRI, gynecologic ultrasonography or pelvis MRI as indicated. Chromosomal analysis was also performed when required. Surgical management such as pelviscopy was considered if secondary dysmenorrhea was suspected on the pelvic examination, the serum CA-125 level and the gynecological ultrasonography. For the diagnosis of AUB, physical examinations, hematologic studies (blood clotting factors, platelets, etc.), hormonal studies and gynecologic ultrasonography were conducted to exclude organic causes, and also a careful history was taken for previous medications such as oral contraceptives. 2.4 Statistical analysis All data are presented as percentage of patients and means ± S.D. Comparisons between groups I and II were made with the use of χ 2 -test and Fisher's exact test. Statistical significance was considered as p values <0.05. 3 Results Among 1280 patients who presented with menstruation-related symptoms, complete clinical data were available for 1237 patients (646 for group I and 591 patients for group II). The age of menarche was 13.0 ± 1.4 years for group I and 13.7 ± 1.4 years for group II. The average ages of the patients on their first visits were 16.7 ± 2.4 years for group I and 24.2 ± 2.8 years for group II. Table 1 shows the menstruation-related problems for each group. AUB (40.9%) and dysmenorrhea (32.5%) were the most frequent presentations for groups I and II, respectively ( Table 1 ). Amenorrhea was the second common problem comprising 29.1% of group I and 31.6% of group II. As shown in Table 2 , primary amenorrhea was classified in hypergonadotropic hypogonadism, hypogonadotropic hypogonadism and eugonadodism ( Table 2 ). The distribution of these three types of amenorrhea differed between groups I and II, and especially for the distribution of hypogonadotropic hypogonadism and eugonadism ( p = 0.0113). For hypergonadotropic hypogonadism, Tuner's syndrome was the most common cause of primary amenorrhea in both groups (group I versus group II, 16.1% versus 12.5%, p = 0.76). For patients with Turner's syndrome, the classical type (58.8%) was the most common, and other types included mosaicism (35.3%) and the variant type (5.9%). Hypogonadotropic hypogonadism was the most common type of primary amenorrhea in group I, and this diagnosis included idiopathic hypothalamic hypogonadism, constitutional delay and endocrine disorders such as congenital adrenal hyperplasia, Cushing's syndrome, hyperprolacinemia and hypothyroisim ( Table 2 ). In group II, two patients were diagnosed as having congenital adrenal hyperplasia, and their subtypes were 21-hydroxylase deficiency and 17α-hydroxylase deficiency (data not shown). Eugonadism was the most common type of primary amenorrhea in group II and the second most frequent type in group I. This type of primary amenorrhea was a significantly higher proportion of group II than of group I ( p = 0.0025), and it included anatomic causes, intersex disorders and inappropriate feedback. The anatomic causes included M-R-K syndrome, vaginal septum and imperforated hymen. M-R-K syndrome was the most frequent anatomic cause of primary amenorrhea in both groups (21.0% versus 50.0%, respectively); however, it was significantly more frequent in group II ( p = 0.0038). There were a total of 29 patients with M-R-K syndrome, and the McIndoe operation was performed for nine patients. Secondary amenorrhea was also classified into three types, the same as primary amenorrhea ( Table 3 ). Chromosomal competent hypergonadotropic hypogonadism was also more common in group II ( p = 0.0036). Among the cases of hypergonadotropic hypogonadism, 3 patients with Turner's syndrome presented with the clinical features of premature ovarian failure (POF). Among the 33 chromosomally competent POF, 4 patients showed positive thyroid related autoantibodies and all these patients belong to group II. There was 1 patient with progressive systemic sclerosis, 1 patient with pelvic tuberculosis and two patients with SLE, and these patients also belong to group II. Among the patients with chromosomal competent ovarian failure, none of them had previous histories of chemotherapy or radiation therapy. Hypothalamic amenorrhea was the most frequent cause of secondary amenorrhea in both groups (49.1% versus 41.4%, respectively, p = 0.11). While weight loss was the more frequent symptom in group I (11.9% versus 5.3%, respectively, p = 0.042), hyperprolactinemia was more frequent in group II (8.8% versus 14.3%, respectively). There was a significantly higher proportion of patients with a serum prolactin level higher than 100 ng/ml in group II (0.8% versus 11.6%, respectively, p = 0.0003), and pituitary adenoma was more common in group II (0.8% versus 9.6%, respectively). Among 32 patients with hyperprolactinemia, 8 were diagnosed with macroadenoma, 8 with microadenoma, and 11 patients were managed surgically. Among the 4 patients with drug-induced hyperprolactinemia who were belonged to group I, 3 patients had mood disorders and 1 patient had schizophrenia, and they were all on either antidepressants or anti-psychotic medication when they first presented to the clinic. Chronic anovulation was the second most common cause in both groups (30.1% versus 21.9%, respectively), but this was without statistical significance between the two groups ( p = 0.11). Fifty-two patients (8.0%) of group I and 50 patients (8.5%) of group II had complaints of infrequent menstruation (>35 days). Of the 52 patients with oligomenorrhoea in group I, 31 patients were diagnosed as PCOS (59.6%) and 4 patients had hyperprolactinemia (7.7%). In group II, 32 patients were diagnosed as PCOS (64%), 4 patients had hyperprolactinemia (8%), and the rest of the causes were of unknown origin. In group I (21.7%) and group II (32.5%) of patients visited our clinic with complaints of excessively painful menstruation. Dysmenorrhea was a common menstruation-related problem in both groups, but it was a significantly more frequent cause of seeking professional care in group II (21.7% versus 32.5%, respectively, p < 0.0001). Although there was no difference in the proportion of primary dysmenorrhea, secondary dysmenorrhea was more common in group II ( p < 0.0001, Table 1 ). Endometriosis was the most frequent cause of secondary dysmenorrhea in both groups (87.0% (20/23) for group I and 77.7% (80/103) for group II). Eleven patients (47.8%) of group I and 45 patients (43.7%) of group II were diagnosed pathologically as having endometriosis. There were 7 patients with uterine myoma and 3 patients with uterine adenomyosis in group II ( Table 4 ). The proportion of endometriosis patients without endometrioma was higher in group I than in group II ( p = 0.042). It was rather unexpected that we did not have any patients with pelvic inflammatory disease as a main cause of secondary dysmenorrhea. Several NSAIDs (non-steroidal anti-inflammatory drugs) such as aceclofenac, mefenamic acid, zaltoprofen and naproxen were used to treat primary dysmenorrhea in both groups I and II. Of 110 patients who were prescribed mefenamic acids as the initial treatment, 97 patients had either improvement or disappearance of symptoms; this was an 88.2% response rate. Aceclofenac, naproxen and zaltoprofen showed response rates of 82.8% (53/64), 81.3% (13/16) and 60.0% (6/10), respectively. Second line medications with other NSAIDs were used when the first line therapy failed, and these treatments were effective in most cases. The overall response rate to NSAIDs was 98.9% (183/185). In group I (40.9%) and group II (26.7%) of patients had complaints of having either long or frequent periods, and then they were clinically diagnosed with AUB; this was more frequent in group I ( Table 1 , p < 0.0001). Dysfunctional uterine bleeding (DUB) was the most frequent cause of abnormal uterine bleeding, followed by PCOS and hyperprolactinemia ( Table 5 ). As Table 5 shows, the order of frequency was similar in both groups, but there was a larger proportion of the DUB patients in group I (81.4% versus 63.3%, respectively, p < 0.0001), and hyperprolctinemia was more prevalent in group II (3.4% versus 12.0%, respectively, p = 0.0005). In group II, AUB related to the endocrinopathies (15.2% versus 25.9%, respectively, p = 0.0064) and the benign lesions of genital tract (0.4% versus 6.4%, respectively, p < 0.0001) were more frequent than in group I. For 186 patients with PCOS, amenorrhea (39.2%), oligomenorrhea (33.9%) and AUB (26.9%) were the common clinical features in order of frequency, and hirsutism was noted in only 6 patients (3.2%). The clinical presentation of PCOS did not show any difference between two groups. The mean free testosterone levels were 1.71 ± 1.23 and 1.60 ± 1.07 pg/ml for groups I and II, respectively, and polycystic ovary was found on pelvic ultrasonography for 49.5% of group I patients and 63.2% of group II patients, and this was without significant difference ( p = 0.059). In group I (0.3%) and group II (0.7%) of patients had visited our clinic for premenstrual syndrome, and there was no significant difference between the two groups ( p = 0.43). 4 Discussion This is one of the few studies that have been conducted on patients with menstruation-related problems from a single institution. The mean age of menarche of 1237 young women who visited our YLC was 13.4 ± 1.5 years, which was similar to 13.2 years of a Korean study of 11,424 teenage girls published in 1996 [9] . However, there still exists a discrepancy compared to the age of menarche (12.8 years) for Caucasian American girls [10] . There are several available classifications of amenorrhea, and Reindollar's classification [11] was used for this study; it was classified as hypergonadotropic hypogonadism, hypogonadotropic hypogonadism and eugonadism. Turner's syndrome was the most common cause of hypergonadotropic hypogonadism type amenorrhea, as was noted in previous studies [6] . The classical type was the most frequent subtype of Turner's syndrome (58.8%), and these patients did not accompany with other anomalies. Hypogonadotropic hypogonadism was the most common type of primary and secondary amenorrhea of group I. Thirteen patients in group I responded to exogenous administration of estrogen–progestin therapy and they acquired secondary sexual characteristics and regular menstruation. For 12 patients diagnosed as constitutional delay (with a mean age of 15.3 years), the mean height was 156.9 cm while the mean height of this total age group was reported to be 159 ± 5.2 cm in a previous study [12] . Also, these patients in our group showed the delayed breast development of Tanner stage II–III. In contrast, the most common type of primary amenorrhea for group II was the eugonadotropic type, and most of them had anatomic causes such as M-R-K syndrome, imperforated hymen and vaginal septum. Among the 29 M-R-K patients, only 1 patient (3.4%) had combined renal anomalies (horseshoe kidney) even though it has been reported that this syndrome was often associated with renal anomalies in 35–70% of these cases, and skeletal anomalies are supposed to be present in 12–50% of these cases [13] . Regarding the secondary amenorrhea, we could reconfirm that chronic anovulation, chromosomal competent ovarian failure and hyperprolactinenia were the three most common causes, the same as Reindollar's study [11] . Chromosomally competent ovarian failure was the more frequent cause of secondary amenorrhea than chromosomal incompetent types in both groups. The proportion of POF patients with normal chromosomes was more prominent in group II ( p = 0.0003). Therefore, chromosomal study should be considered for POF patients less than 20 years of age. Turner's syndrome was the most common chromosomal abnormality in both groups (1.6% versus 0.6%, respectively). Many of the Turner's syndrome women lose all of their germ cells before birth, and some of them, as our patients, could have enough germ cells remaining at puberty to initiate a part of the pubertal process and have regular, cyclic menses during at least the early period of their adolescence. As the 45X karyotype may have undetected Y DNA, we evaluated the SRY sequences to detect the possible risk of gonadal tumor; however, SRY was not detected in any of our patients. POF represented 6.4% of primary amenorrhea patients and 11.7% of secondary amenorrhea patients in our study; these were similar results to previous studies in which POF represented 10–28% of primary amenorrhea patients and 4–18% of secondary amenorrhea patients [14–16] . Moreover, hypergonadotropic hypogonadism may also be related to endocrine dysfunction of the thyroid and adrenal gland. The most common cause of POF is autoimmune in nature, and these patients have an increased risk of developing other autoimmune abnormalities such as Hashimoto thyroiditis, adrenal insufficiency and pernicious anemia. It has been reported that among the thyroid autoantibodies, the positive rate for microsomal antibodies was 27.5 and 12.5% had thyrogloblulin antibodies [17] , and also in this study, 21.1% (4 patients) had positive thyroid related autoantibodies. As our study shows, most of the adolescents and premarital women with secondary amenorrhea showed low to normal levels of FSH and LH ( Table 3 ). Hypothalamic amenorrhea with non-specific causes and chronic anovulation were the most common causes of secondary amenorrhea for both groups. There was a higher proportion of patients with weight loss induced secondary amenorrhea in group I (11.9% versus 5.3%, respectively, p = 0.042), and this weight loss was related to a pathological obsession with a thin self-image. As normal menstruation is resumed in over 80% of the cases once the causative factor such as stress, weight loss or eating disorder was removed [18] , treatment should focus on correcting the underlying problems rather than amenorrhea itself. Hyperprolactinemia was the main cause of secondary amenorrhea rather than primary amenorrhea; moreover, it was a much more common cause in group II patients who had already passed the pubertal process. This was probably caused by the fact that an increased production of estrogen during early puberty initiates the elevation of prolactin [19] . Oligomenorrhea showed similar proportions in both groups (8.0% versus 8.5%, respectively). As a medical treatment, periodic medroxyprogesterone or cyclic pill medication was given to protect the endometrium from hyperplasia. Life style modification such as weight reduction and regular exercise was also recommended. For oligomenorrheic patients with menses at less than an 8-week interval, no therapy was given except for reassurance and regular follow-up. Women 40–90% of reproductive age suffer from dysmenorrhea [20–22] and this interferes with daily activities in 10–12% of these women [23] . The prevalence of dysmenorrhea has been shown to increase steadily among menstruating adolescents, from 39% at age 12 years to 72% at age 17 years [24] . Among adolescents with dysmenorrhea, 15% are incapacitated for 1–3 days monthly, and 30–50% of them complain of mild pain [25] . Even though we could not evaluate the incidence of dysmenorrhea in the general population, this study showed that secondary dysmenorrhea was relatively more common than primary dysmenorrhea in group II (3.6% versus 17.4%, respectively, p < 0.0001). All the patients having primary dysmenorrhea that interfered with their daily activities were recommended to use NSAIDs such as aceclofenac, naproxen and mefenamic acid at the onset of menstruation, and to continue the medication for 1–3 days according the patient's pain duration. Almost 90% of patients experienced pain relief within 3 months. Although oral contraceptives (OCP) can be used as a primary treatment for sexually active girls and this treatment has shown a 95% response rate [26] , it was not used as an initial treatment in our center and it was only used as a second line therapy. It is well known that the single most common cause of secondary dysmenorrhea and chronic pelvic pain during adolescence is endometriosis. In group II, the number of patients with pathologically confirmed endometrioma and the number of patients with only pelvic endometriosis spots were similar. However, in group I, there were more patients with endometriosis without endometrioma ( p = 0.042). These findings suggested that diagnostic pelviscopy might be necessary even in the absence of any positive ultrasonographic findings of endometriosis for patients who do not respond to the medical treatment, and especially for the adolescent patients. In sexually active adolescents with dysmenorrhea, pelvic examination with collection of cervical samples for gonorrhea, chlamydia and Papanicolau smear were performed. But we had no cases confirmed to have PID as a main cause of dysmenorrhea. Uterine bleeding more than 7 days, more than 6 pads daily or with intervals less than 21 days defines abnormal uterine bleeding. In our study, the two groups showed slightly different etiologies for AUB. DUB was 81.4% of the AUB in group I, but DUB was less frequent in group II (63.3%, p = 0.00003). Anovulatory bleeding associated with H-P-O immaturity occurs frequently during first 2–3 years after menarche, and it usually takes up to 2–5 years till there is stabilization of the H-P-O axis in adolescents. McDonough and Gantt [27] observed anovulation in 55–82% of adolescents between menarche and 2 years postmenarche, 30–55% from 2 to 4 years postmenarche, and 20% from 4 to 5 years postmenarche. In this study, the rate of DUB in group I was high during first 2 years from menarche (89/97, 91.8%), and it decreased after 5 years from menarche (50/76, 65.8%). DUB became a less frequent cause of AUB as time progressed from menarche ( p < 0.001). In addition, adolescents with earlier menarche are reported to develop ovulatory cycles earlier than those adolescents with a later onset of menarche [27] . Therefore, the later onset of menarche in Korea might affect the incidence of DUB. In group II, organic causes such as uterine fibroid and endometrial pathology ( p < 0.001), and endocrinopathies ( p = 0.006) such as hyperprolactinemia and PCOS were more frequently associated with AUB than was noted for group I ( Table 5 ). The clinical features of PCOS were variable, and of the 186 patients diagnosed as PCOS, 73 patients (39.2%) showed amenorrhea and 63 patients (33.9%) showed oligomenorrhea. The incidence of hirsutism in our study was only 3.2%, which was much less common than has been reported in western countries (25–75.5%) [28–30] . 5 Conclusion In summary, menstruation-related problems constitute an important area of the reproductive health services and more attention should be given to diagnosis and treat menstrual complaints. Early diagnosis and the appropriate treatment of menstrual symptoms are important for adolescent and premarital women for the prevention of possible adverse effects on their reproductive potential. Therefore, it is very important that the medical community should keep abreast with changes for the management of menstruation-related problems for all types of patients. Although this study has some limitations for generalizing our results to the overall population, we believe that this study does reflect, at least in part, the current menstruation-related problems of adolescents and premarital women in Korea. References [1] G.B. Slap Menstrual disorders in adolescence Best Prac Res Clin Obstet Gynecol 17 2003 75 92 [2] J.R. Klein I.F. Litt Epidemiology of adolescent dysmenorrhea Pediatrics 68 1981 661 664 [3] A. Ziv J.R. Boulet G.B. Slap Utilization of physician offices by adolescents in the United States Pediatrics 104 1999 35 42 [4] G.M. Jagiello H.A. Kaminetsky P. Ricks Jr. R.J. Ryan Primary amenorrhea. A cytogenetic and endocrinologic study of 18 cases JAMA 198 1966 148 156 [5] F. Pettersson H. Fries S.J. Nillius Epidemiology of secondary amenorrhea. Incidence and prevalence rates Am J Obstet Gynecol 117 1973 80 86 [6] R.H. Reindollar J.R. Byrd P.G. McDonough Delayed sexual development: a study of 252 patients Am J Obstet Gynecol 140 1981 371 380 [7] K. Munster P. Helm L. Schmidt Secondary amenorrhoea: prevalence and medical contact-a cross-sectional study from a Danish county Brit J Obstet Gynaecol 99 1992 430 433 [8] M.E. Herman-Giddens E.J. Slora R.C. Wasserman Secondary sexual characteristics and menses in young girls seen in office practice: a study from the Pediatric Research in Office Settings network Pediatrics 99 1997 505 512 [9] J.C. Shin C. Lee J. Moon Menarche in Korean adolescent girls Kor J Obstet Gynecol 39 1996 865 879 [10] World Health Organization Task Force on Adolescent Reproductive Health World Health Organization multicenter study on menstrual and ovulatory patterns in adolescent girls. I. A multicenter cross-sectional study of menarche J Adolesc Health Care 7 1986 229 235 [11] R.H. Reindollar M. Novak S.P. Tho P.G. McDonough Adult-onset amenorrhea: a study of 262 patients Am J Obstet Gynecol 155 1986 531 543 [12] C.H. Hong Textbook of pediatrics 8th ed. 2004 Korean Textbook Publishers Seoul [13] L.S. Timmereck R.H. Reindollar Contemperary issues in primary amenorrhea Obstet Gynecol Clin North Am 30 2003 287 302 [14] C.B. Coulam S.C. Adamson J.F. Annegers Incidence of premature ovarian failure Obstet Gynecol 67 1986 604 606 [15] C. Jamin Premature ovarian failure syndrome Contracept Fertil Sex 23 1995 153 156 [16] A.R. LaBarbera M.M. Miller C. Ober R.W. Rebar Autoimmune etiology in premature ovarian failure Am J Reprod Immunol Microbiol 16 1988 115 122 [17] L. Falsetti S. Scalchi M.T. Villani G. Bugari Premature ovarian failure Gynecol Endocrinol 13 1999 189 195 [18] R.B. Perkins J.E. Hall K.A. Martin Aetiology, previous menstrual function and patterns of neuro-endocrine disturbance as prognostic indicators in hypothalamic amenorrhoea Hum Reprod 16 2001 2198 2205 [19] S.J. Emans M.R. Laufer D.P. Goldstein Pediatric and adolescent gynecology 4th ed. 1998 Lippincott-Williams & Wilkins Philadelphia [20] B. Andersch I. Milsom An epidemiologic study of young woman with dysmenorrhea Am J Obstet Gynecol 144 1982 655 660 [21] O. Widholm Dysmenorrhea during adolescence Acta Obstet Gynecol Scand 87 Suppl. 1979 61 66 [22] A.S. Coco Primary dysmenorrhea Am Fam Physician 60 1999 489 496 [23] W.Y. Zhang A. Li Wan Po Efficacy of minor analgesics in primary dysmenorrhoea: a systematic review Brit J Obstet Gynaecol 105 1998 780 789 [24] L.S. Neinstein Menstrual problems in adolescents Med Clin North Am 74 1990 1181 1203 [25] J. Teperi M. Rimpela Menstrual pain, health and behaviour in girls Soc Sci Med 29 1989 163 169 [26] P.S. Simmons Common gynecologic problems in adolescents Prim Care 15 1988 629 642 [27] P.G. McDonough P. Gantt Dysfunctional bleeding in the adolescent B.N. Barwin S. Belisle Adolescent gynecology and sexuality 1982 Masson Publishing New York