Introduction

Female sexual function is influenced by physical, psychological, social and cultural factors. Sexual dysfunction has been defined as a disturbance in any phases of sexual response (e.g., desire/excitement, orgasm, sexual satisfaction) or the presence of pain during sexual intercourse causing personal distress or interpersonal challenges [1]. The prevalence of female sexual dysfunction is high, ranging from 38 to 85.2 % [2, 3]. Aging increases the prevalence of sexual dysfunction; however, the transitional phase after the menopause contributes to female sexual dysfunction regardless of chronological age [3, 4]. The main changes that occur in sexual function after menopause are decreased sexual desire, reduced vaginal lubrication, anorgasmia and dyspareunia [5].

Some authors have postulated different mechanisms by which pelvic floor muscle (PFM) strength may influence female sexual function. Kegel suggested that PFM weakness could contribute to the inability of a woman to achieve orgasm [6]. According to Shafik, an increase in the strength of the muscles attached to the corpus cavernosum of the clitoris could lead to increased arousal and orgasm [7]. Graber and Kline-Graber found a significantly lower pubococcygeus muscle strength in women who had anorgasmia compared with women who had orgasms [8]. Another study that included women with primary complaints of sexual dysfunction showed that moderate and strong PFM strength was associated with the highest Female Sexual Function Index (FSFI) scores [9]. However, others have found no association between sexual function and PFM strength [10, 11].

Studies have shown that after menopause, there may be decreased function of the PFM due to estrogen deficiency [12, 13]. Although postmenopausal status can be considered a risk factor for the development of sexual dysfunction, there is scant knowledge on the relationship between sexual function and PFM strength. This study aimed to evaluate the relationship between PFM strength and sexual function in postmenopausal women. The relationship between reported urinary incontinence (UI) and sexual function was also evaluated.

Materials and methods

This was a cross-sectional clinical study approved by the Research Ethics Committee of the School Health Center of the Ribeirão Preto Medical School, University of São Paulo (CSE-FMRP-USP) under protocol no. 259/CEPCSE-FMRP-USP. All women recruited for the study gave written informed consent to participate.

Participants

The study was advertised on local radio and at the School Health Center of the Ribeirao Preto Medical School at the University of São Paulo, Ribeirão Preto City. Women interested in participating in the project were asked to contact an assistant researcher to receive more information about the study at the School Health Center of the Ribeirao Preto Medical School. After receiving information and agreeing to be evaluated for eligibility screening, women who fulfilled all the inclusion criteria were included in the study.

For inclusion the women had to be heterosexual and to have been postmenopausal for a maximum of 10 years, with the criterion for postmenopausal status being cessation of menstrual cycles for more than 12 months (World Health Organization), and to be currently sexually active with intercourse. Sexual intercourse was defined as penile penetration of the vagina in the previous 4 weeks [14, 15]. Exclusion criteria were diabetes mellitus and any reported thyroid disease, intolerance of or discomfort during the examination to evaluate PFM strength, allergy to gel or latex condoms, prolapse greater than stage 1, and inability to contract the PFM as assessed by vaginal palpation.

Measurements

The ability to contract the PFM was first evaluated by digital palpation. PFM strength was evaluated using vaginal manometry (PeritronTM, Cardio-Design, Lara Victoria, Australia). Peritron has been found to have good intrarater and moderate interrater reliability [16, 17]. Before PFM measurement, the woman completed the FSFI [15] and the International Consultation on Incontinence Questionnaire-Urinary Incontinence Short Form (ICIQ-UI SF) [18]. The examiner of PFM strength was blinded to the results of the FSFI and the ICIQ-UI SF. The FSFI is a self-reported 19-item questionnaire validated for the Portuguese language [15]. It covers six domains of sexual function: lubrication, arousal, desire, pain, orgasm, and global satisfaction. The total scores range between 2 and 36. The domain scores and total scores are associated with better sexual function. The use of cut-off values on the FSFI enables the identification of women with sexual dysfunction [11]. A total score of 26.5 or less is considered to indicate sexual dysfunction [19]. The ICIQ-UI SF is used to evaluate the symptoms, severity and impact of UI on quality of life. The questionnaire has been validated in the Portuguese language and shows good test–retest reliability [20]. The third question (prevalence of UI) on the ICIQ-UI SF was used to identify the presence of UI. UI was classified into four severity levels on the basis of the mean questionnaire scores according to the system proposed by Klovning et al. [21]: mild (score 1 – 5), moderate (score 6 – 12), severe (score 13 – 18), and very severe (score 19 – 21) [21]. Question 6 was used only to classify the type of UI.

The PFM was assessed by one physiotherapist at the Laboratory of Functional Evaluation of the Pelvic Floor, Ribeirão Preto Medical School. The participants were informed about PFM anatomy and function using diagrams. Digital palpation was performed with the woman in the supine position with semiflexed hips and knees. Circular closing around the palpating fingers with movement in a cranial ventral direction was considered a correct contraction [16]. Five minutes after vaginal palpation, the probe of the Peritron™ (Cardio-Design, Lara Victoria, Australia) covered with a latex condom coated with aqueous lubricating gel was inserted above the level of the hymenal ring to the full extent of the compressible portion until 1 cm of the sheath remained visible outside the vaginal introitus [17]. Three maximum voluntary contractions (MVC) were requested, with a 30-s interval between each contraction. In all measurements, the inward movement of the probe and the perineum were used as an indication that the device was recording PFM contraction correctly [16]. Cocontraction of the hip adductors and the gluteus muscle was discouraged by requesting the woman to perform the Valsalva maneuver. The mean pressure of three MVCs was used in the analysis of PFM strength.

Statistical power and analysis

The sample size to detect a difference of 8 cmH2O using the Peritron™, with a standard deviation of 15, was calculated as 108. For allocation of subjects to each group we considered the prevalence of sexual dysfunction [22] assuming a proportion around 2.5 participants with sexual dysfunction for each participant without sexual dysfunction. A statistical power of 80 % and alpha 0.05 were used (G*Power power analysis program, version 3.1.7).

The data were analyzed using SAS® software, version 9.2. The relationship between the two quantitative variables (FSFI score and ICIQ-UI SF score) was quantified using Spearman’s rank correlation coefficient (ρ). Values between 0.10 and 0.29 were considered to indicate a weak correlation, values between 0.30 and 0.49 a moderate correlation, and values between 0.50 and 1 a strong correlation [23]. Fisher’s exact test was used to compare the following variables between the groups with and without sexual dysfunction: age, body mass index (BMI), marital status, number of vaginal births, number of nulliparous women, and number of women with UI complaint. The Pearson’s chi-squared test was used to compare the use of hormone therapy between the groups. The Mann–Whitney test for independent samples was used to compare PFM strength (MVC pressure) between women with and without sexual dysfunction.

Results

Of 154 postmenopausal women recruited, 39 were not eligible because they had not had sexual intercourse in the previous 4 weeks, and 2 were not included because they were not able to contract their PFM. Thus a total of 113 women fulfilled the eligibility criteria and were included in the study. Table 1 shows the demographics of the women with and without sexual dysfunction. There were no statistically significant differences between the groups in any of the variables.

Table 1 Comparison of the characteristics of the study participants with and without sexual dysfunction
Full size table

In women with sexual dysfunction the mean FSFI scores for all domains were significantly lower than in women without sexual dysfunction (Table 2). PFM strength was significantly lower in woman with sexual dysfunction than in women without sexual dysfunction (Table 2). The median ICIQ-UI SF score was 4 (range 0 – 69.2) for the whole sample (n = 113). Table 3 shows the type and severity of UI symptoms in women with and without sexual dysfunction. There was no significant difference in the severity of UI between the two groups (p = 0.07). There was a negative weak correlation between the total ICIQ-UI SF score and the FSFI domain scores (Table 4).

Table 2 Domain scores on the FSFI and PFM strength (maximum voluntary contraction pressure) in women with sexual dysfunction (n = 82) and without sexual dysfunction (n = 31)
Full size table
Table 3 Type and severity of urinary incontinence symptoms in women with and without sexual dysfunction
Full size table
Table 4 Correlation between the ICIQ-UI SF score and the FSFI domains and total score for the whole sample of 113 women
Full size table

Discussion

The main finding of the present study was that women without sexual dysfunction had stronger PFM. UI prevalence and severity were similar between women with and without sexual dysfunction and there was a weak negative correlation between UI severity and sexual function. It has been estimated that around 50 % of postmenopausal women have some complaints of sexual dysfunction, with the most common being hypoactive sexual desire, dyspareunia and anorgasmia [5]. The prevalence of sexual dysfunction found in this study was high, but is in agreement with previously reported rates in the range 38 % to 85.2 % [24]. In this study, the FSFI scores for all domains were lower in women with sexual dysfunction. The desire domain median score of less than 3 in the group with sexual dysfunction indicates a specific risk of hypoactive sexual desire disorder. A desire domain score of less than 3 is an established cut-off value to differentiate women with and without this condition [25].

It has been postulated that the PFM may play an important role in sexual function. The pubococcygeus and ileococcygeus muscles are responsible for involuntary contractions during orgasm [9]. Some authors have stated that an increase in PFM strength could lead to better involuntary contraction of the PFM and to increased arousal and orgasmic response [6, 7]. In the present study, women with sexual dysfunction had weaker PFM than those without sexual dysfunction. However, as the present study was cross sectional, it is not possible to establish a cause–effect relationship. A systematic review of the literature on the effects of PFM training on female sexual function did not demonstrate a clear relationship between PFM strength and sexual function [26]. In most of the eight RCTs included in this review this analysis was not performed. Conflicting results were found between one study that did not show an association between PFM strength (Brink score) and sexual function [27] and one study that showed a medium correlation between changes in sexual function and PFM strength (manometry) [28].

Few other previous cross-sectional studies have specifically evaluated the relationship between PFM strength and sexual function. Lowenstein et al. [9] conducted a retrospective chart review of 176 women with primary complaints of sexual dysfunction recruited in a tertiary urogynecological clinic who were evaluated using the FSFI (47 % had libido dysfunction, 40 % anorgasmia, 8 % dyspareunia and 5 % other complaints) [9]. Women with moderate to strong PFM had higher scores in the orgasm and arousal domains. However, the study used a scale that has not been validated for assessment of PFM strength. In a study including 40 nulliparous women, Martinez et al. found an association between higher scores on the desire domain and total FSFI scores and higher PFM strength. Although they used a vaginal manometer to evaluate muscle strength, they did not report reliability and validity data for the device. In addition, for analytical purposes, the PFM contractions measured with the perineometer were arbitrarily classified as strong or weak using a cut-off value above and below 8.8 cmH2O [29]. The mean age of the women included in the two studies discussed above was lower than of those participating in the present study, and Martinez et al. did not report the prevalence of UI or its association with sexual function [29]. Lowenstein et al. found that 45 % of women with weak PFM had UI compared with 25 % of those with moderate to strong PFM.

UI is a distressing symptom for postmenopausal women [30]. In the present study, a large number of women reported UI. Several studies have shown a higher prevalence of sexual dysfunction in women with UI [10, 31]. However, this relationship frequently disappears when confounding variables, such as menopause itself, lack of desire and pain, are controlled for. The prevalence of reported UI of 48.1 % in women with sexual dysfunction found in the present study is consistent with the prevalence range of 19 % to 50 % reported in the literature [31]. We found no significant difference in ICIQ-UI SF scores between women with and without sexual dysfunction. There was a small negative correlation between the desire, arousal, satisfaction, pain and total FSFI scores and the total ICIQ-UI SF score. This is in agreement with the findings of other studies, which have also shown lower scores in the desire, satisfaction and pain domains of the FSFI in women with incontinence than in women without incontinence [31].

Severe UI has frequently been correlated with sexual dysfunction [11]. However, most of the women in this study reported moderate UI, and this may have contributed to the weak correlation between sexual function and the impact of UI on quality of life. There was no statistically significant difference between the groups regarding severity of UI, possibly because of the small number of women with each level of severity. These results should therefore be interpreted with caution, and more research is warranted in this area.

The present study had some limitations and strengths that need to be addressed. We acknowledge that the tool used to evaluate sexual function is a screening tool used as a potential aid in the diagnosis of sexual dysfunction and does not allow a complete diagnosis of sexual dysfunction. Moreover, we did not evaluate distress and we did not have any information about the use of drugs that may have influenced sensory function. We cannot discard the possibility that women with hormone alterations leading to anovulation were included, as we used the World Health Organization criteria for postmenopause that do not include hormone levels of follicle-stimulating hormone. Women unable to contract their PFM were excluded, and these women could have had the weakest muscles, but we consider that this would not have interfered with our results as only two such women were excluded. Our study design does not allow any statement to be made as to cause and effect, and other uncontrolled variables could also have influenced sexual function in the postmenopausal women studied. However, as far as we are aware, this study is one of only a few investigating the relationship between PFM strength and sexual dysfunction in women. The study used reliable and valid instruments to evaluate both sexual function and PFM strength. In addition, a single experienced examiner conducted all the examinations.

In conclusion, this study showed that postmenopausal women without sexual dysfunction have a stronger PFM than women with sexual dysfunction. There was a weak correlation between UI severity and sexual function. Further randomized controlled trials are needed to establish the cause–effect relationships between PFM strength and sexual dysfunction and UI in postmenopausal women.