Coitus and Menstruation in Perimenopausal Women

Abstract

The objective of the study was to determine whether coitus during menstruation in perimenopausal women is associated with increased menstrual flow defined as either heavier flow, more days of flow, or both.

Initial interviews for the Stanford Menopause Society Study were conducted locally with 160 perimenopausal women. Six weeks later, additional data were gathered from 121 women. At a third interview, 6-10 weeks later at Stanford University, 56 of 57 subjects (mean age 48.8 + 3.9 SD) provided data on menstrual flow and coital behavior.

At both the first and third interview, more than half of the women reported a pattern of increased menstrual flow. At the third interview, 83% of those reporting increased menstrual flow also reported a behavioral pattern of coitus during menses. Only 10% with diminished flow acknowledged such a pattern. Women who abstained from sex during menses were not less regular in coital activity at other times nor did the data reveal them to be less orgasmic than women who did not abstain.

Our data reveal an association between coitus during menstruation and longer and/or heavier menstrual flow. The etiology of such increased flow is usually believed to be hormonal, but coitus during menstruation may be another important variable.ent 5

Introduction

Population statistics suggest that in the USA more than 12 million women are currently within their perimenopausal transition years. Since 19791-3 it has been increasingly recognized that on average the 7 years before menopause - more recently identified as the perimenopausal transition - are characterized by a wide range of changes in menstrual patterns. In Treloar's major work3, 25,000 menstrual cycle years of data were prospectively recorded and analyzed, following three generations of women. Treloar showed the broad pattern of variation in menstrual cycle length.

The reproductively fertile years are characterized by relative stability in cycle length and a relative trend toward a 29.5-day cycle. In contrast, the first 7 years and the last 7 years of menstrual flow are characterized by a wide variation in the cycle length. Women are likely to have extremely long or extremely short cycles or some combination of the two. It is precisely this sudden break in the stable pattern of cycling that serves to indicate that the menopausal transition has begun3.

Since Treloar's work, other studies have affirmed the relationship between abnormally long and short cycles and perimenopausal status among women after age 52,4. Some women experience increases in the days of menstrual flow while others flow for fewer days. Some flow more heavily, while other have a scantier flow. Such changes in menstrual flow during the transition period are thought to signal the impending menopause and to be related to changes in sex hormone levels.

The potential relationship of sexual behavior to uterine function was a relatively new consideration in 1979, when the longitudinal study of this report was initiated. Masters and Johnson suggested the possibility of increased contractility during orgasmic response5,6. Although their work suggested such a mechanism in women, it had been preceded by work with animals. A number of mammalian species have shown uterine contractions coincident with and shortly after copulation (reviewed in refs 7-9). There were observations of an increased myometrial motility during human menstruation10. Increased impedance of uterine arterial blood flow peaking at the onset of menstruation has also been reported11. However, none of these studies has examined the relationship of these uterine variables to coital activity.

In one study of 498 consecutive female infertility patients those who acknowledged coitus during menses showed a significantly higher incidence of endometriosis than patients who said they abstained from coitus during menses. Both occasional and frequent coitus during menses showed a positive association with the incidence of endometriosis while such coital activity was unrelated to pelvic inflammatory disease12.

A review of our data as well as the literature led us to explore whether perimenopausal women who engaged in coitus during menstruation were more likely to experience heavier or longer duration of menstrual flow. Considering the uterine physiology, altered uterine contractions during menstruation could lead to a more forceful expulsion of menses from the uterine cavity as well as retro-grade menstruation. The increase in the duration of flow may be a reflection of the slow regeneration of the endometrium caused by these contractile forces. Since increased menstrual flow is a recurrent problem in the menopausal transition years, the Stanford Menopausal Study population offered a logical population in which to explore this question.

 

METHODS

Procedure

Recruitment of subjects followed a 5-min television news interview or similar radio or newspaper announcement in which the senior author invited women in the San Francisco Bay area to participate in a study of the menopause. The study was described as a way for women to contribute to knowledge about healthy women as they approach and reach the menopause. Although the principle aim of the Stanford Menopause Study was to gather data about female sexuality and hormones during the menopausal transition period, we were careful to dilute awareness of this goal by collecting data on a wide range of topics throughout the early months of the study including the data reported here.

At every interview, women filled out a questionnaire and also engaged in face-to-face interviews as part of a protocol that included a sequence of standardized tests such as blood pressure measurements. At both the first and the third interview, women were asked about the kinds of changes they had observed globally in their menstrual flow patterns as they entered and moved through the menopausal transition period. For menstrual flow: was there more blood, less, mixed patterns, or no change? For duration of menstrual flow: were there fewer days, more days, variable, or no change? At the first interview, several (3%) had not menstruated in several months as shown in Table 2. At the second interview, women completed a sexual history questionnaire that had Likert-scaled questions on a variety of sexual elements that included frequency of sexual arousal and orgasm during coitus, overall satisfaction with their sexual relationship, use of fantasy, masturbation, and so forth.13 Sexually-active women were invited to a more time consuming and demanding third interview that subjects were told would focus on sexual life. Interview 3 was conducted at the senior author's office at Stanford University where each was interviewed and had a blood sample drawn.13,14

Subjects

A total of 205 presumably healthy women were enrolled in response to the original call for subjects. One hundred and sixty gave informed consent and completed the initial interview with a research assistant. Of these 153 (96%) were perimenopausal as evidenced by symptoms of: hot flashes, and/or night sweats or a noticeable change in menstrual cycle length, flow or pattern (see refs 13-16 for previous reports from this study). These data were collected 8 years before the first publication of normative follicle stimulating hormone values as perimenopausal transition markers.17 Women were neither hysterectomized nor ovariectomized, and had not used contraceptive or replacement hormones within the prior 6 months. Of these, 121 women completed the second interview and 57 completed the third interview. While women were not asked to explain why they did not continue in the study, attrition was probably due to the necessity of having a blood sample drawn, the amount of time involved in participating in the study, and because the third interview involved coming to Stanford University rather than meeting a researcher locally. In addition, it was at this point that subjects were told that the study would now focus on sexuality; for many reasons, including celibacy, this may have affected their willingness to continue.

At the third interview, 56 women provided data on menstrual flow pattern and coital behavior during menstruation: two were menopausal - 16 and 22 months from their last cycle respectively - and provided retrospective data. All subjects were Caucasian and had a mean (+SD) age of 48.8(+3.9). Subjects had a mean (+SD) age at menarche of 12.8 years (+1.3), age at first coitus of 20.9(+3.9), age at first child of 23.6 years(+3.6), number of pregnancies of 2.8(+1.7), number of live births of 2.2(+1.3), and years of education of 15.2(+2.4). Data for relationship status revealed that 71.4% were married, 21.4% were divorced, and 7.1% were never married. All but three subjects had a sex partner at the time of the third interview, and the average length of ongoing relationships was 18.4 years(+10.4).

<--Prev                    Next-->

Results

Menstrual flow changes

At the first interview, 153 women completed questionnaires, and 121 of these women provided data on menstrual flow change. Of these, 115(75%) reported globally noticeable changes in the menstrual flow, with 81 women noting some form of increase and 34 women some form of decrease. Thirty-two women did not answer the question. These data are presented in Table 1.

Table 1-- Number of women with increased (+), decreased (-), and no change (=) in number of days and amount of menstrual flow at the menopausal transition
	Menstrual flow
	Days	Amount	Number of women
Increased flow	+	+	13
	+	=	8
	=	+	43
	spotting		10
	=	+	6
	broken healing start and stop		1
Subtotal	81			(53%)
Decreased	-	-	3
	-	=	4
	=	-	27
Subtotal	34			(22%)
No Change	=	=	6
Subtotal	6			(4%)
No data	32			(21%)
Total	153			(100%)

 

Changes in the length of the cycle were also reported by a large proportion of these women. (Menstrual cycle length is defined as the time span from day 1 of a cycle to the next day 1, regardless of duration or quantity of flow. Thus a `missed period' by definition by definition is equivalent to a long cycle.) Table 2 shows that 81% of the 144 women who answered the question were aware of specific changes in cycle length coincident with their other perimenopausal symptoms (hot flashes, night sweats and so forth). As shown in Table 2, of 144 women, 40% noticed an increase in cycle length, 17% noticed their cycles had become shorter and 24% noticed a change but sometimes this was experienced as shorter cycles and other times as longer cycles.

Table 2
Number and percentage of women with a change in menstrual cycle length at the menopausal transition.

Table 2 Number and percentage of women with a change in menstrual cycle length at the menopausal transition.
	n	%
Longer	58	40
Shorter	24	17
Mixed	35	24
No Change	22	15
Stopped > 3 months ago	5	3
Total	144	99

 

 

Sexual behavior and menstrual flow

Interview 3 specifically focused on details of the overall perimenopausal menstrual flow pattern to allow clearer discrimination between cycle length, menses duration, and amount of flow with patterns of sexual behavior. Each woman was asked `Do you engage in sexual intercourse when you are menstruating?' Table 3 shows the individual data that relate menstrual flow patterns to sexual behavior during menstruation. The data were grouped according to menstrual flow patterns with arrows designating changes in both the duration and the amount of flow: increasing, decreasing or unchanged.

These data show that every woman who reported both increased amount and increased duration of flow answered `yes' to coitus during menstruation, whereas every woman who reported both decreased amount and decreased duration of flow answered `no'.

Overall, 83% of women who said they experienced increased menstrual flow during the menopausal transition period reported `yes' to the behavior pattern of coitus during menstruation whereas 90% of women with lighter and or/shorter duration of menstrual flow reported abstention. Of 13 subjects with a mixed pattern of bleeding change (one flow variable increased, the other decreased), 61% reported `yes' and 39% reported that they abstained. These data are presented in Table 4 and are highly significant.

Table 4

Table 4 Change in menstrual flow and number of women having coitus during menses
	Coitus during menses
Menstrual flow change	Yes	No
Increased	19	4
Mixed Change	8	5
Decreased	2	18
Total	29	27

Among the 56 women, six cases, identified in Table 3 with asterisks next to their code numbers, were more closely considered because of their comments. These six subjects commented `rare' or `minimal' to the question concerning coitus during menstruation making their answer of `yes' or `no' ambiguous. In an additional test these six subjects were deleted and the analyses were repeated without them. The results retain their significance, x2 (1,n=50)=24.7, p<0.005

Overall, 83% of women who said they experienced increased menstrual flow during the menopausal transition period reported `yes' to the behavior pattern of coitus during menstruation whereas 90% of women with lighter and or/shorter duration of menstrual flow reported abstention. Of 13 subjects with a mixed pattern of bleeding change (one flow variable increased, the other decreased), 61% reported `yes' and 39% reported that they abstained. These data are presented in Table 4 and are highly significant. Among the 56 women, six cases, identified in Table 3 with asterisks next to their code numbers, were more closely considered because of their comments. These six subjects commented `rare' or `minimal' to the question concerning coitus during menstruation making their answer of `yes' or `no' ambiguous. In an additional test these six subjects were deleted and the analyses were repeated without them. The results retain their significance, x2 (1,n=50)=24.7, p<0.005.

As estradiol levels were available (see ref.15), we assessed whether estradiol was different in the three menstrual flow groups (increased, mixed, decreased). No significant difference was seen between the mean level of estradiol (+SD) in subjects with increased (57.3 pg/ml+38.4), mixed (71.7 pg/ml+39.6), or decreased (61.0pg/ml+41.8) menstrual flow, F(2,43)=<1. Mean age in the three groups was also examined. Similarly, mean ages (+SD) of subjects with increased (48.6+4.0), mixed (48.2+4.2), and decreased (49.4+3.6) menstrual flow patterns did not differ significantly.

Women who acknowledged coitus during menses were compared to those who abstained to test whether they differed in other aspects of their sexual behavior. All but three subjects were currently sexually active; one woman had been having coitus during menstruation until a recent cessation of sexual activity (#85). Regularity of coitus was dichotomized in accord with our previous studies as Weekly (W) for those who never missed a non-menstruating week or Sporadic (S) for those sexually active women whose prospectively recorded behavioral calendar data showed less than weekly coitus (see refs. 14,15). Regularity of coitus was determined using calendar data for a 4-week period just prior to interview 4; this interview occurred 6-10 weeks after interview 3. These data are shown in Table 3.

First, we tested whether there were differences in regularity of coitus (Weekly vs. Sporadic) between these two groups. Results showed no significant difference, x2(1,n=49)=2.81,p>0.10<0.50. The women who said they abstained from coitus during menses were not significantly more apt to be sporadic in their sexual activity than those who acknowledged coitus during menstrual flow.

Next, we addressed whether women who refrained from coitus during menstrual flow reported a higher or lower frequency of orgasm than those who did not. Answers to how often they experienced orgasm during coitus on a 7-point Likert scale (1=never to 7=everytime) were assessed. Results showed no significant difference in mean (+SD) frequency of orgasm score between those who did not engage in coitus during menstrual flow (5.4+1.9) and those who did (4.9+2.0).

Finally, we assessed whether reported frequency of orgasm was different in the three menstrual flow groups (increased, mixed, decreased). There were no significant differences for mean (+SD) frequency of orgasm between those who reported increased (4.8+2.1), mixed (5.6+ 1.2), or decreased (5.5+1.8) menstrual flow.

<--Prev                    Next-->

Discussion

The data in this study show a fourfold elevation in increased menstrual flow among those perimenopausal women who reported a pattern of coitus during menses compared to those who did not (65.5% and 14.8% respectively). They also show that women who reported that they refrained from coitus during menstruation had nine times the incidence of diminished flow compared to those who did not refrain from coitus when they were menstruating (62.9% and 6.9% respectively).

Previously, researchers reported that increased menstrual flow - whether heavier flow or prolonged length of flow - was relatively frequent; it occurred in about half of women as they approached menopause3. Findings in our study were similar in both samples (see Tables 1 and 3). Because absolute quantity of flow could not be measured, this study relied on the perception of the study subjects. However, for many of the subjects in the Stanford Menopause Study, this increased amount and duration of menstrual flow was clearly perceived and disturbing. Likewise, women described their own perception of changes in cycle length and these data were not always collected prospectively. The fact that subjects in this study recorded each day of menses on daily calendars may have contributed to their awareness of cycle length. The problems of recurrent heavy menstrual flow during the menopausal transition are now recognized as a common phenomenon1,2 and are believed to be endocrine in origin.

The observations here suggest that if our subjects' retrospective self-reports accurately reflect their pattern of sexual behavior, then coitus during menses may modify menstrual flow. An alternative hypothesis is that women who were more active sexually or more orgasmic may have been more likely to have coitus when they were menstruating. Our data do not support this hypothesis as there was no significant difference in regularity of coital activity (sporadic vs. weekly) between those who abstained from coitus during menstrual flow and those who did not. Nor did frequency of orgasm show a significant difference between groups. Because women who had coitus during menses engaged in coitus during additional days of their cycle compared to women who abstained , they may have had a higher frequency of coitus. Thus frequency of coitus might also be an important variable.

Other possible explanations for our data are that women who menstruated for longer durations may have been less reluctant to abstain from coitus during menses, but this explanation would apply solely to women with increased duration and not to those with increased flow only. Moreover, women might have engaged in coitus during menses because they believed that coitus would contract the uterus and staunch the blood flow. Our data suggest that, at least for perimenopausal women, this belief has no validity.

However, a women's pattern of coital activity has previously been shown to affect other menstrual and endocrine parameters. Delayed age of first coitus was associated with subsequent infertility.20 Women who had coital activity at least once a week when they were not menstruating (regular, 'weekly') showed an increased frequency of normal length menstrual cycles21,22, higher levels of luteal phase estrogen23, and a higher frequency of fertile type basal body temperature graphs23,24 than women who were sporadically active or abstaining.

Because coitus during menses may alter physiological events affecting menstruation in perimenopausal women, the next question concerns a putative mechanism. In order to relate sexual behavior to uterine blood flow, it is useful to consider the physiology of menstruation. This was quite elegantly detailed by Bartelmez in 193718 and further described in 195919. These classic papers reported the characteristic sloughing of the endometrium as it passed out through the cervix and vagina during menstruation. The subsequent endometrial "micro-cuts" that formed from broken blood vessels then underwent clotting and regeneration. Bartelmez described the nature of menstrual flow - 1 day of light flow with one day of heavier flow - which usually alternated twice. Finally, sequential physiological processes during the 1 to 4 days of waning menstruation concluded the menstrual sequelae.

The literature shows that there are various factors that contribute to uterine contractility during sexual activity. Among these are orgasm, arousal, oxytocin, the prostaglandins and semen, or some other unidentified factor or substance. Are the changes in menstrual flow secondary to the thrusting effects of the penis, the effect of the ejaculate, or some other factor that triggers the contractions of the uterus? The role of orgasm was initially addressed in 1991 in experimental studies involving the stumptail macaque9. Female orgasm appeared to be the trigger for uterine contractions in this species. Sexually related uterine contractions occurred with or without the presence of penile intromission, provided the female showed evidence of sexual climax through the acceleration of her heart rate and subsequent species-typical facial expressions. Dominant females showed more frequent sexual climaxes during copulation that lower ranked females. Thus, not every copulatory event led to uterine contractions. The climactic uterine contractions in these macaques were distinguished from an initial telemetrically recorded uterine response of lesser magnitude than occurred to penile thrusting alone. The telemetrically recorded pattern of "climactic contractions" of the uterus was more powerful (higher amplitude) and continued for several minutes after the penis was withdrawn before returning to baseline. Both copulation leading to orgasm and orgasm in homosexual pairings without penetration produced uterine climactic type contractions.

The work of Masters and Johnson5,6 suggests a similar potential in women. Moreover, both abstaining and non-abstaining women in our study reported orgasm to be a commonly occurring experience at coitus. Likewise those with heavier flow were no more orgasmic than those with lighter flow. The behavior of coitus during menstruation appears to be the relevant issue.

Increases in heavy flow and the increases in uterine contractility could provide an adverse combination leading to retrograde menstruation. This may be an important consideration regarding the etiology of endometriosis and would support the findings of Filer and Wu12. Admittedly, published data on the incidence of intercourse during menses appear to be very limited. The only reference located was Filer and Wu's and they reported that slightly over half of the sample of 498 women acknowledged coitus during menses frequently or occasionally. The rest of the women they studied said they abstained12. Our data were similar with 48% saying that they abstained. Other causes of menstrual anomalies are well recognized; these include thyroid dysfunction, drug use and psychogenic pathology. However, because ours was a study of healthy women these factors are not relevant.

Could coital activity during menses increase uterine contractions? If so, these accentuated uterine contractions during menstruation could stimulate the expulsive forces that could be interpreted by the woman as increased heaviness of flow. Mechanical explanations of such an effect seem plausible. Heaviness of flow, combined with uterine contractility, can both express the menses as well as push them into the peritoneal cavity. It is less likely that the increased blood flow is due to an endometritis because the endometriosis/menstrual-coital activity data of Filer and Wu do not support an increase of pelvic inflammatory disease in women who acknowledged coitus during menses12. Lengthening of menstrual flow associated with coitus during menses might be a result of the effects of the contractile forces at orgasm.

Our data as well as those of Filer and Wu12 show that an increased frequency with which women report experiencing sex during menses does not parallel the extent of pathology. Both those who frequently , as well as those who occasionally, reported engaging in coitus during menses were substantially more likely to present with increased menstrual flow at perimenopause in our study and endometriosis in theirs. It may be that the perception of abstention is a clear one while the perception of "occasional menstrually occurring coitus" is subject to the need for better quantification in future studies.

Oxytocin is a known hormonal by-product when nipple suction is sufficient to elicit milk during lactation. Vaginal-cervical stimulation is a potent stimulus for oxytocin secretion25. In 1994, Carmichael and colleagues reported that plasma oxytocin levels in women and men rose in close temporal relationship to orgasm induced through masturbation26.

The precise mechanisms involved need elucidation. However, the relative hypermenorrhea associated with coitus during menses, as well as the increased incidence of endometriosis12 point to potential mechanical processes that exacerbate menstrual flow. This could support the philosophy expressed by the Moslem-Judaic Codes that proscribed coitus during menstruation. Data on the flow habits of these groups have not been published.

If coitus is exacerbating menstrual flow during the time of life when women are vulnerable to hysterectomies because of heavy menstrual flow, this new information might be useful in reducing the incidence of such surgery. Likewise, if coitus during perimenopausal menses produces such effects, it is reasonable to study the same question in younger women. Moreover, further investigations into perimenopausal hypermenorrhea should yield more complete information. We encourage other investigators to gather and examine similar data in a variety of populations.

A printed copy of this article is available in most university libraries but can also be ordered through Athena Institute
A bibliography of Dr. Cutler's Published Work

Reproduction in whole or in part in any form or medium without express written permission of Athena Institute is prohibited.

<--Prev