1 School of Public Health, University of California, Berkeley, California, 2 School of Public Health and Community Medicine, University of Washington, Seattle, Washington, 3 Department of Obstetrics and Gynecology, University of Milan School of Medicine, Milan, Italy and 4 Department of Obstetrics and Gynecology, University of Wisconsin Medical School, Madison, Wisconsin, USA
5 To whom correspondence should be addressed at: School of Public Health, University of California, 140 Warren Hall, Berkeley,CA 94720-7360, USA. Email: eskenazi{at}calmail.berkeley.edu
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: epidemiology/leiomyoma/menstrual cycle/transvaginal ultrasound
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
There are plausible biological reasons for an association between leiomyoma and menorrhagia and other menstrual cycle characteristics. Leiomyomata gradually increase the volume and surface area of the uterus; this increase expands the endometrium, thus requiring increased time or flow rate to shed during menses. Leiomyomata, especially submucosal ones, may distort the endometrium, causing ulceration or necrosis. The compression of nearby veins in the interior uterine layers can also cause congestion of blood flow in the endometrium, with increased releases (Koutsilieris, 1992; Lumsden and Wallace, 1998
; Vollenhoven, 1998
; Hickey and Fraser, 2000
; Robboy et al., 2000
).
Most studies of the relationship of leiomyoma to menorrhagia and other menstrual cycle characteristics have been based on care-seeking populations, including women undergoing myomectomy or hysterectomy (Buttram, 1986; Cramer et al., 1995
; Sato et al., 2000
; Faerstein et al., 2001
) or tubal ligation (Chen et al., 2001
), with mixed results. However, clinic-based studies may not be generalizable, because women who have menstrual irregularities may be more likely than asymptomatic women to seek treatment and to be referred for surgery, and women who seek sterilization are more likely to be fertile than other women.
To date, only one study has examined the association of leiomyoma and menstrual cycle changes in a non-care-seeking population. In a random sample of >900 health plan members, the presence of ultrasound-diagnosed leiomyoma was associated with gushing-type bleeding (Wegienka et al., 2003). In a general population of non-care-seeking Italian women, we previously reported that dyspareunia and non-cyclic pelvic pain, but not dysmenorrhoea, increased in severity with the presence of uterine leiomyomata (Lippman et al., 2003
). The purpose of the present study was to examine in these women the relationship between leiomyoma presence, number, volume, location and position and menstrual cycle characteristics (cycle length, flow length and heaviness of flow). These women were recruited based on where they resided not on medical treatment or location of medical care.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
All women 2050 years old (n=751) and those between 50 and 60 years old who were still menstruating (n=14) were invited to undergo a pelvic examination and transvaginal ultrasound following an interview; 662 (87%) agreed to participate and were examined. Only women with an intact uterus who were currently menstruating were eligible for analysis, leading to the exclusion of women who had previous hysterectomy (n=29), who had not menstruated in the previous year (n=25), were using oral contraceptives (n=119) or intra-uterine devices (IUDs) (n=34), were pregnant (n=1) or were lactating (n=5), leaving 451 women (some women had more than one reason for exclusion). Preliminary analysis showed that none of the 110 women under 30 years of age had leiomyoma, so these younger women were excluded from further analysis. The final study sample therefore consisted of 341 currently menstruating women, who were 30 years or older.
Procedures
Study participants were interviewed by trained nurse-interviewers. In addition to being asked about their reproductive and life histories, participants were asked the following four questions about their menstrual cycle characteristics: (i) In the past year, how long have your cycles been?; (ii) In the past year, were your cycles regular, i.e. was the number of days from the start of one period to the start of the next about the same, give or take 4 days?; (iii) In the past year, on average, how many days did your flow last?; and (iv) During the past year, how would you describe the heaviness of your menstrual flow: scanty, moderate or heavy? From the responses to these questions, we derived continuous variables for menstrual cycle and menstrual flow length in days, and a categorical variable for menstrual flow heaviness (scanty, moderate or heavy).
Following the interview, a transvaginal ultrasound was performed by a gynaecologist at the Mangiagalli Hospital of the University of Milan or at the Desio Hospital. The procedure used the Aloka colour Doppler 680 and SSD 2000, 5 MHz transvaginal probe for imaging, and 6 MHz pulsed Doppler system for blood flow analysis. Each ultrasound examination was videotaped and reviewed by a second gynaecologist for quality control.
Based on the ultrasound examination, the following leiomyoma variables were created for each woman: (i) presence of one or more leiomyoma; (ii) total number of leiomyomata; (iii) leiomyoma tissue layer location, i.e. any subserosal, any submucosal or any intramural leiomyoma (three separate indicators were created because multiple leiomyomata could be present and a single leiomyomata could span more than one location); (iv) leiomyoma axial position (four variables), i.e. presence of any leiomyoma in the fundus, isthmus, anterior or posterior; and (v) dimension, i.e. volume in cm3 for each myoma using the formula for a prolate ellipse (West and Lumsden, 1989; Schwartz et al., 1998
). For women with multiple leiomyomata, three summary volume measures were created: mean volume, maximum volume and total summed volume.
Statistical analysis
Because the analyses use cross-sectional data, the associations between leiomyoma and menstrual cycle characteristics could be modelled with either as the dependent variable. For the associations with continuous variables, cycle length and flow length, we designed these variables as dependent and the leiomyoma characteristics as independent. For ease of interpretation for associations with the three-category variable, heaviness of flow, we designated the leiomyoma characteristics as the dependent variable and heaviness of flow as the independent variable. Bivariate associations were analysed with 2, ANOVA and Fisher's exact tests.
For multivariate analyses, we used least squares mean regression for cycle length and flow length, and logistic regression for heaviness of flow. Covariates considered included: age at interview, education, consumption of caffeinated beverage, cigarette smoking habits, alcohol consumption, history of IUD and oral contraceptive use, age at menarche, body mass index (BMI), gravidity, parity and history of previous diagnosis of endometriosis, leiomyoma, ovarian problems or other reproductive problems (Samadi et al., 1996; Schwartz, 2001
). Covariates that confounded the relationship between menstrual cycle characteristics and presence of fibroids, as determined by a 10% change in the parameter estimate, were kept in the model. All analyses employed model-free standard errors, which are valid even under departures from standard assumptions (Huber, 1967
). Women who reported irregular cycles were excluded from analyses of cycle length. Analyses of leiomyoma volume, tissue location and axial position were restricted to the subset of women with leiomyomata. Data analysis was conducted using Stata Statistical Software, Release 8.1.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
Menstrual cycle characteristics and presence of leiomyoma
The relationship of menstrual cycle characteristics to leiomyoma is shown in Table II. In crude analysis, having a leiomyoma was not associated with menstrual cycle length (P=0.40) or flow length (P=0.90) and remained unrelated after adjusting for covariates (age, BMI, smoking, parity, duration of oral contraceptive use or past IUD use). In the crude analysis, having leiomyoma was associated with scanty menstrual flow [odds ratio (OR) =2.4, 95% confidence interval (CI) 1.25.1] and heavy menstrual flow (OR =1.8, 95% CI 1.03.1) relative to moderate flow, but, after adjusting for covariates, these relationships remained somewhat elevated but were no longer significant (OR for scanty flow =1.9, 95% CI 0.84.3; OR for heavy flow =1.3, 95% CI 0.72.5). Adjusting for TCDD exposure did not alter these results.
|
The median of the mean volume per fibroid was 4.4 cm3 (25th, 75th percentiles = 1.4, 10.2). The median of the summed volumes was 6.6 cm3 (25th, 75th percentiles =1.9, 19.3). None of the measures of leiomyoma volume (mean, maximum or summed) varied by menstrual characteristics including cycle length (P>0.6, all comparisons), menstrual flow length (P>0.2, all comparisons) or flow heaviness (P>0.6, all comparisons) (data not shown).
Menstrual cycle characteristics and leiomyoma tissue layer location and axial position
A summary of menstrual cycle characteristics by leiomyoma tissue layer location (subserosal or intramural) and axial position (anterior or posterior) is presented in Table III. None of the menstrual characteristics varied by location or position in crude (shown in Table III) or age-adjusted analyses (data not shown). There were only a few leiomyoma cases in submucosal (n=3) or pedunculated (n=3) tissue layer locations and fundal (n=4) or isthmal (n=5) axial positions; therefore, no statistical analyses were performed with these locations or positions.
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In our study, the presence, number and volume of leiomyomata were unrelated to menstrual cycle length, flow length and heaviness of flow. Our findings are consistent with previous studies (Cramer et al., 1995; Sato et al., 2000
; Faerstein et al., 2001
), but inconsistent with others (Clevenger-Hoeft et al., 1999
; Chen et al., 2001
; Wegienka et al., 2003
). Similarly, although some studies (Akkad et al., 1995
; Clevenger-Hoeft et al., 1999
), but not all (Wegienka et al., 2003
), have suggested that tissue layer location and the axial position of leiomyomata may play a role in severity of menstrual symptoms, we did not find an association between menstrual cycle characteristics and location (subserosal or intramural) or position (anterior or posterior) of leiomyomata.
Our findings probably differ from previous studies of care-seeking populations because women with menstrual irregularities are more likely to seek treatment. However, our results also differ somewhat from the only other population-based study of women sampled from a health maintenance organization. In that study, Wegienka et al. (2003) found that women with leiomyomata were more likely to report gushing-type bleeding (defined as an episode of bleeding which could not be controlled by pads or tampons) than women without leiomyomata. This definition may capture women with more problematic menstrual bleeding than we were able to in our study by asking about usual heaviness of menstrual flow over the past year with choices of scanty, moderate and heavy. Nevertheless, we found a significant crude association not only with heavy flow but also with scanty flow, which diminished after control of potential confounders. We note, however, that somewhat more women with leiomyoma reported using maxi pads than women without leiomyoma (P=0.07), lending some support to an association of leiomyoma and heavy flow.
Our study had several strengths. As a population-based cross-sectional study, the results are generalizable to the population of similar women, at least in Italy. The questionnaires were administered by highly trained nurse-interviewers, and the examinations were performed by experienced gynaecologist/ultrasonographers for research purposes. The participation rate in SWHS was high (81%). However, our study also had limitations. As noted elsewhere, if women with the most severe symptoms are more likely to be treated with hysterectomy, then we would have underestimated not only the overall prevalence of leiomyomata, but also the association between leiomyomata and heavy menstrual flow. However, of the 29 women excluded from ultrasonography because of prior hysterectomy, women who had hysterectomies because of leiomyomata were no more likely to report bleeding as an indication for surgery than women who had hysterectomies for reasons other than leiomyomata (26.9 versus 33.3%, 2=0.06, P=0.8).
Our study was limited to Caucasians, who may be at lower risk for leiomyoma (Kjerulff et al., 1996; Baird et al., 2003
). Thus, we could not examine differences in menstrual characteristics associated with leiomyoma by race. It is possible that population differences may exist in the genetics (Kitawaki et al., 2001
; Sato et al., 2002
) and other biological aspects of leiomyoma which might affect menstrual characteristics (Roth et al., 2003
).
In conclusion, based on this population-based study of non-care-seeking Caucasian women, we found little evidence that abnormalities in menstrual cycle length, flow length and heaviness of flow were related to number, volume, tissue layer location (subserosal or intramural) and axial position (anterior or posterior) of leiomyomata. The association of leiomyomata with menstrual cycle abnormalities has been based principally on previous studies of clinic-based populations, particularly of women seeking hysterectomy. Because asymptomatic women are under-represented in such populations, the association of leiomyoma with menstrual cycle disorders in the general population may previously have been overestimated.
![]() |
Acknowledgements |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Baird DD, Dunson DB, Hill MC, Cousins D and Schectman JM (2003) High cumulative incidence of uterine leiomyoma in black and white women: ultrasound evidence. Am J Obstet Gynecol 188, 100107.[CrossRef][ISI][Medline]
Borgfeldt C and Andolf E (2000) Transvaginal ultrasonographic findings in the uterus and the endometrium: low prevalence of leiomyoma in a random sample of women age 2540 years. Acta Obstet Gynecol Scand 79, 202207.[CrossRef][ISI][Medline]
Buttram VC, Jr. (1986) Uterine leiomyomaaetiology, symptomatology and management. Prog Clin Biol Res 225, 275296.[Medline]
Chen CR, Buck GM, Courey NG, Perez KM and Wactawski-Wende J (2001) Risk factors for uterine fibroids among women undergoing tubal sterilization. Am J Epidemiol 153, 2026.
Chryssikopoulos A and Loghis C (1986) Indications and results of total hysterectomy. Int Surg 71, 188194.[ISI][Medline]
Clevenger-Hoeft M, Syrop C, Stovall D and Van Voorhis B (1999) Sonohysterography in premenopausal women with and without abnormal bleeding. Obstet Gynecol 94, 516520.
Cramer SF, Horiszny JA and Leppert P (1995) Epidemiology of uterine leiomyomata. With an etiologic hypothesis. J Reprod Med 40, 595600.[ISI][Medline]
Eskenazi B, Mocarelli P, Warner M, Samuels S, Vercellini P, Olive D, Needham L, Patterson D and Brambilla P (2000) Seveso Women's Health Study: a study on the effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin on reproductive health. Chemosphere 40, 12471253.[CrossRef][ISI][Medline]
Faerstein E, Szklo M and Rosenshein N (2001) Risk factors for uterine leiomyoma: a practice-based case-control study. I. African-American heritage, reproductive history, body size and smoking. Am J Epidemiol 1, 110.
Farquhar CM and Steiner CA (2002) Hysterectomy rates in the United States 1990-1997. Obstet Gynecol 99, 229234.
Graves WP (1933) Uterine myomata (fibroids). In Curtis, A.H. (ed.) Obstetrics and Gynecology.WB Saunders, Philadelphia, Vol. II, pp. 745832.
Hickey M and Fraser IS (2000) Clinical implications of disturbances of uterine vascular morphology and function. Baillieres Clin Obstet Gynaecol 14, 937951.[CrossRef]
Huber P (1967) The behavior of maximum likelihood estimates under non-standard conditions. In Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, Vol. 1. University of California Press, Berkeley, CA, pp. 221233.
Kitawaki J, Obayashi H, Ishihara H, Koshiba H, Kusuki I, Kado N, Tsukamoto K, Hasegawa G, Nakamura N and Honjo H (2001) Oestrogen receptor-alpha gene polymorphism is associated with endometriosis, adenomyosis, and leiomyoma. Hum Reprod 16, 5155.
Kjerulff KH, Langenberg P, Seidman JD, Stolley PD and Guzinski GM (1996) Uterine leiomyomata. Racial differences in severity, symptoms and age at diagnosis. J Reprod Med 41, 483490.[ISI][Medline]
Koutsilieris M (1992) Pathophysiology of uterine leiomyomata. Biochem Cell Biol 70, 273278.[ISI][Medline]
Lepine LA, Hillis SD, Marchbanks PA, Koonin LM, Morrow B, Kieke BA and Wilcox LS (1997) Hysterectomy surveillanceUnited States, 19801993. Morbidity and Mortality Weekly Report. CDC Surveillance Summaries 46, 115.
Lippman S, Warner M, Samuels S, Olive D, Vercellini P and Eskenazi B (2003) Uterine fibroids and gynecologic pain symptoms in a population-based study. Fertil Steril 80, 14881494.[CrossRef][ISI][Medline]
Lumsden MA and Wallace EM (1998) Clinical presentation of uterine fibroids. Baillieres Clin Obstet Gynaecol 12, 177195.[ISI][Medline]
Luoto R, Kaprio J, Keskimaki I, Pohjanlahti JP and Rutanen EM (1994) Incidence, causes and surgical methods for hysterectomy in Finland, 19871989. Int J Epidemiol 23, 348358.[Abstract]
Robboy SJ, Bentley RC, Butnor K and Anderson MC (2000) Pathology and pathophysiology of uterine smooth-muscle tumors. Environ Health Perspect 108, 779784.[ISI][Medline]
Roth TM, Gustilo-Ashby T, Barber MD and Myers ER (2003) Effects of race and clinical factors on short-term outcomes of abdominal myomectomy. Obstet Gynecol 101, 881884.
Samadi AR, Lee NC, Flanders WD, Boring JR, 3rd and Parris EB (1996) Risk factors for self-reported uterine fibroids: a case-control study. Am J Public Health 86, 858862.[Abstract]
Sato F, Miyake H, Nishi M, Mori M and Kudo R (2000) Early normal menstrual cycle pattern and the development of uterine leiomyomata. J Womens Health Gend Based Med 9, 299302.[CrossRef][ISI][Medline]
Sato F, Mori M, Nishi M, Kudo R and Miyake H (2002) Familial aggregation of uterine myomas in Japanese women. J Epidemiol 12, 249253.[Medline]
Schwartz LB, Rutkowski N, Horan C, Nachtigall LE, Snyder J and Goldstein SR (1998) Use of transvaginal ultrasonography to monitor the effects of tamoxifen on uterine leiomyoma size and ovarian cyst formation. J Ultrasound Med 17, 699703.
Schwartz SM (2001) Epidemiology of uterine leiomyoma. Clin Obstet Gynecol 44, 316326.[CrossRef][ISI][Medline]
Treloar SA, Do KA, O'Connor VM, O'Connor DT, Yeo MA and Martin NG (1999) Predictors of hysterectomy: an Australian study. Am J Obstet Gynecol 180, 945954.[ISI][Medline]
Velebil P, Wingo PA, Xia Z, Wilcox LS and Peterson HB (1995) Rate of hospitalization for gynecologic disorders among reproductive-age women in the United States. Obstet Gynecol 86, 764769.
Vollenhoven B (1998) Introduction: the epidemiology of uterine leiomyomata. Baillieres Clin Obstet Gynaecol 12, 169176.[ISI][Medline]
Wegienka G, Baird DD, Hertz-Picciotto I, Harlow SD, Steege JF, Hill MC, Schectman JM and Hartmann KE (2003) Self-reported heavy bleeding associated with uterine leiomyoma. Obstet Gynecol 101, 431437.
West CP and Lumsden MA (1989) Fibroids and menorrhagia. Baillieres Clin Obstet Gynaecol 3, 357374.[ISI][Medline]
Submitted on March 28, 2003; resubmitted on March 5, 2004; accepted on June 16, 2004.
|