Aging Effects on Pelvic Floor Support: A Pilot Study Comparing Young versus Older Nulliparous Women

Abstract

Introduction and Hypothesis:

We sought to determine age-related changes to the pelvic floor in the absence of childbirth effects.

Methods:

A case-control study was conducted from June 2017–August 2018 comparing two groups of nulliparous women: <40 years old and ≥70 years old. Clinical evaluation included POP-Q, instrumented speculum testing, and handgrip strength. Dynamic 3D-stress MRI was performed on all women to obtain genital and levator hiatus (LH) lengths, LH area, and levator bowl volume. LH shape was quantified using a novel measure called the “V-U Index.” Pubovisceral muscle (PVM) cross-sectional area (CSA) was also measured. Bivariate comparisons between the two groups were made for all variables. Effect sizes were calculated for MRI measures.

Results:

Twelve young and nine older nulliparous women were included. Levator bowl volume at rest was 83% larger in older women (108.0 ± 34.5 cm³ vs 59.2 ± 19.3 cm³, p=.001, d=1.82). MRI genital hiatus at rest was larger among the older group (2.7 ± 0.6 cm vs 3.5 ± 0.6 cm, p=.007, d=1.34). V-U index, a measure of LH shape where 0 = “V” and 1 = “U,” differed between groups—indicating a more “U”-like shape among older women (0.71 ± 0.23 vs 0.35 ± 0.18, p=.001, d=1.72). Handgrip strength was lower in the older vs young group (23.2 ± 5.2 N vs 33.4 ± 5.2 N, p<.0001); however, Kegel augmentation force and PVM CSA were similar (3.2 ± 1.1 N vs 3.3 ± 2.2 N, p =.89, and 0.8 ± 0.3 cm² vs 0.7 ± 0.2 cm², p=.23, respectively).

Conclusions:

Levator bowl volume at rest was over 80% larger among older women, reflecting a generalized posterior distension with age.

Brief Summary:

Aging is associated with a significant increase in levator bowl volume, which is 80% larger among older (≥70 years) versus young (<40 years) nulliparous women.

Introduction

Over 200,000 women per year undergo surgery due to pelvic organ prolapse, with the highest rates of surgery among women >70 years old [1]. Aging is one of the strongest risk factors for prolapse, and it is also associated with increased severity of prolapse [2]. Despite this knowledge, we have no evidence-based explanation for why this association exists.

Aging research shows that by 80 years of age, approximately 50% of skeletal muscle fibers are lost and that muscle strength also significantly decreases [3]. The levator ani are the skeletal muscles that help establish and maintain pelvic organ support. The pelvic floor, formed by the levators, can be thought of as a “bowl” with varying depth and volume that is interrupted on one side by the urogenital hiatus. One hypothesis to explain the association between aging and prolapse is that aging causes weakening of the levators, which leads to “sagging,” or deepening, of the levator bowl. This, in turn, creates increased space in the pelvis, allowing for descent of the pelvic organs.

Studies attempting to quantify the effect of aging on pelvic floor support are often confounded by vaginal parity, another significant risk factor for prolapse [4,5]. Few studies using nulliparous cohorts exist, and those that do are limited in scope—focusing on either clinical measurements of pelvic floor function or imaging analyses of structural changes [6–8]. The interaction between structural and functional age-related changes in living women remains poorly understood. Therefore, the goal of this pilot study was to combine clinical and functional measurements, as well as novel imaging parameters, to perform a comprehensive analysis of age-related changes to pelvic floor support in living women in the absence of childbirth-related effects. Specifically, we sought to test the hypotheses that aging is associated with: a) smaller and weaker levator ani muscles; b) larger genital hiatus; and c) larger levator bowl volume; and to assess the relative strength of each association. Furthermore, we sought to assess the correlation between handgrip strength, which is a marker of frailty [9,10], and pelvic floor muscle strength.

Materials and Methods

We conducted a case-control study of young (<40 years) and older (≥70 years) nulliparous women from June 2017 to August 2018. The study received approval from the University of Michigan Institutional Review Board (HUM00132937) and all women provided written informed consent. Age criterion for the young group was 18–39 years and ≥70 years for the older group. Women had to be fluent in reading and speaking English. Exclusion criteria included current or past pregnancy past 20 weeks, severe arthritis or condition resulting in impaired hand function, prior hysterectomy, current or past treatment for pelvic organ prolapse, current symptoms of prolapse, and any contraindication to having an MRI.

Women underwent clinical examination that included: Pelvic Organ Prolapse Quantification (POP-Q) Examination [11], instrumented speculum testing to evaluate pelvic floor muscle strength, and assessment of handgrip strength. The POP-Q and instrumented speculum testing were done in low lithotomy position at an inclination of 45 degrees. In addition to the standard POP-Q measurements, genital hiatus and perineal body lengths at rest were also collected. The instrumented vaginal speculum is a device designed to measure isometric intravaginal closure force at rest and during maximal Kegel contraction. Kegel augmentation force is the difference between the two measures [12]. Handgrip strength was assessed using an adjustable-handle dynamometer and the study participant’s dominant hand. In accordance with standardized arm and hand positioning for hand strength measurements, each subject was seated with her shoulder adducted and neutrally rotated, elbow fixed at 90°, forearm in neutral position, and wrist between 0° and 30° dorsiflexion [13]. The average force of three maximal hand contractions was used in analyses.

All women underwent dynamic 3D-stress MRI imaging in the supine position using a multiple Valsalva technique as previously described [14]. Proton-density weighted fast spin-echo imaging was performed in the axial, sagittal, and coronal planes using a 3-Tesla (3-T) Ingenia MRI scanner (Philips Medical Systems, Best, The Netherlands). Figure 1 shows the measures obtained at rest and during maximal Valsalva in the mid-sagittal plane: genital hiatus (inferior pubic point to perineal body), cervix height (vertical distance from horizontal reference line [PICS line] to cervix), and pelvic floor depth (distance from PICS line to perineal body) [15]. Mid-paravaginal height, vaginal length and mid-vaginal width were measured using techniques previously described by Chen et al [16]. Mid-paravaginal height was defined as the distance from the PICS line to the mid-lateral anterior vaginal wall. This measure was similar bilaterally; therefore, only the average was reported. Vaginal length was measured in the mid-sagittal plane from urethral meatus to anterior fornix. Mid-vaginal width was the distance between the lateral vaginal sidewalls at the mid-point of the anterior vaginal wall.

Figure 1. Mid-sagittal MRI measures obtained at rest and during maximal Valsalva.

Pubic symphysis (PS), uterus (Ut), genital hiatus (GH), and perineal body (PB) are labeled. Pelvic inclination correction system (PICS) uses a standardized reference line 34° caudal to the sacrococcygeal to inferior pubic point (SCIPP) line to control for pelvic inclination during Valsalva. Asterisk (*) marks the location used to measure mid-paravaginal height. For the figure, this is shown in the mid-sagittal plane; however, the actual measures were obtained from the lateral margins of the anterior vaginal wall in the para-sagittal planes.

Figure 2 shows 2D area and 3D volume measures that were made at both rest and maximal Valsalva (only resting measures shown in figure). Borders of the mid-sagittal levator area (left panel) were defined by the urogenital hiatus anteriorly, sacrococcygeal to inferior pubic point (SCIPP) line superiorly, and inside of the levator plate (LP) and top of the anal sphincter complex (ASC) inferiorly, as previously described [17]. 3D levator volumes (middle panel) were made using the technique described by Nandikanti et al [18]. The right panel shows the PICS LH area, which was defined by the area inside the levators in the plane of the PICS line. This approach was used in order to standardize the plane in which LH area is measured relative to the bony pelvis. To obtain this measure, the axial plane was rotated until it was parallel to the PICS line plane (i.e. 34° from the SCIPP line, see Figure 1) and the outer border of the levator ani muscles was traced at the level where both pubic bones were visible. The PICS LH area was then divided into a pubic portion, bordered by the margins of the pubic rami, and a posterior portion, bordered by the levators.

Figure 2. MRI levator areas and volume comparing a young (top row) versus older (bottom row) nulliparous woman.

Left panel shows mid-sagittal levator area bordered by the genital hiatus (GH), sacrococcygeal to inferior pubic point (SCIPP) line, levator hiatus (LH), levator plate (LP), and anal sphincter complex (ASC). PS-pubic symphysis. Middle panel shows 3D levator bowl volume. Right panel shows the PICS levator hiatus area divided into pubic and posterior portions. U- urethra, V-vagina, R-rectum.

During our analysis of the PICS LH area images, we observed a notable difference in levator hiatus shape between young and older women, with the young women having more of a “V” shape and the older women having a “U” shape. In order to quantify this difference, we developed a measure called the V-U index, which is shown in Figure 3. This measure was performed on resting images and in the same plane used for the PICS LH area, previously described. To start, two lines were drawn to form a V from the origin of the levator on each side of the pubic bones to the midpoint of the levators behind the rectum. Next, the posterior portion of the levator hiatus area was outlined. The V-U index—the degree to which the shape varied from a perfect V—was calculated as the area outside the triangle divided by the area inside the triangle. The triangle area was used so that variations in overall pelvic size would not affect this measure. Index values closest to 0 indicate a “V”-shaped levator hiatus and those closest to 1 indicate a “U”-shaped levator hiatus.

Figure 3. Method for calculating V-U Index and comparison between a young (top row) and older (bottom row) nulliparous woman.

For ease of understanding, axial images are displayed; however, the V-U Index was actually measured in the PICS plane (same plane in which PICS LH area was measured). Left panel: axial MRI with pubic symphysis (PS), urethra (U), vagina (V), rectum (R), and anal sphincter complex (ASC). Notice the “V” shape of the levators in the young woman and the “U” shape in the older woman. Middle panel: at the pubic symphysis, a V is traced from the origin of the levators on each side to the midpoint of the levators behind the rectum. The posterior portion of the levator hiatus area is outlined. Right panel: The V-U index is calculated—index values closest to 0 indicate a “V”-shaped levator hiatus and those closest to 1 indicate a “U”-shaped levator hiatus.

Pubovisceral muscle (PVM) cross-sectional area (CSA) was measured from MR images using the previously described technique by Masteling et al [16,19]. A 3D multi-echo Dixon sequence was used to quantify fat content in the pubovisceral muscle [20].

Basic demographics and medical, surgical, and family history were collected. Participants also completed the following validated questionnaires: the Pelvic Floor Distress Inventory short form (PFDI-20) [21], the Bristol Stool Scale [22], and the World Health Organization Quality of Life-BREF Questionnaire (WHOQOL-BREF) [23]. The PDFI consists of 20 questions in three domains: pelvic organ prolapse distress inventory (POPDI-6), urinary distress inventory (UDI-6), and the colorectal and anal distress inventory (CRADI-8). The WHOQOL-BREF is a 26-item questionnaire that addresses domains of physical health, psychological health, social relationships, and environment. Scores for each domain range from 4–20, with higher scores indicating better quality-of-life.

Bivariate comparisons between groups were made using student’s t-test for continuous variables, with the exception of POP-Q values, which were compared using Mann-Whitney U. P-values and effect sizes were calculated for comparisons of continuous variables. Cohen’s D (d) was used to determine effect size and the magnitude of mean difference was categorized into the following groups: small (d=0.20), medium (d=0.50), large (d=0.80), very large (d=1.10), and extremely large (d≥1.40) [24,25]. Intraclass Correlation Coefficient (ICC) was calculated to determine interrater reliability for the V-U index measure. A post-hoc power analysis was performed using resting levator bowl volume. Based on the pilot data, a sample size of nine per group would be needed to detect a statistical difference between groups with 90% power. Statistical analyses of the questionnaire data were conducted using Stata, version 15 (StataCorp LP, College Station, TX). All other statistical analyses were conducted IBM SPSS Statistics software version 24.0 (copyright 2018 IBM Corporation, Armonk, NY).

Results

Twenty-one women were recruited for the study; 12 in the young group and nine in the older group. Table 1 shows basic demographic data and clinical measurements from the two groups. On average, women in the older group were 50 years older than those in the young group. All women in the older group reported having undergone menopause and 11/12 young women reported being premenopausal. One young participant did not answer this question. As expected, average handgrip strength was 29% lower in the older group; however, pelvic floor strength as measured by Kegel augmentation force, was similar between groups. The correlation (r) between Kegel augmentation force and handgrip strength was 0.32 (p=.16). There were differences in POP-Q measures of Level III support, with point Bp in the older group being 1 cm lower than in the young group, and the young group having a larger genital hiatus (GH)—both resting and straining. Median increase in GH size from rest to strain was 0.5 cm for both groups (0.5 IQR (0, 1.0) for both groups, p=.85).

Table 1.

Demographics and clinical measurements in young and older nulliparous women without prolapse symptoms

Demographics and Clinical Measurements	Young (n=12)	Older (n=9)	p-valuea
Age, years	24.2 ± 3.2	74.1 ± 4.7	<.0001
Caucasianb	9 (75.0)	7 (77.8)	>.99
Parity	0	0	N/A
Body Mass Index, kg/m2	25.9 ± 4.3	29.6 ± 8.0	.23
Handgrip Strength, N	31.5 ± 5.5	22.3 ± 5.2	.001
Kegel Augmentation Force, N	3.3 ± 2.2	3.2 ± 1.1	.89
POP-Qc
Ba	−3.0 (−3.0, −2.0)	−2.0 (−3.0, −1.5)	.13
C	−7.0 (−8.5, −6.0)	−7.0 (−6.5, −8.0)	.88
Bp	−3.0 (−3.0, −3.0)	−2.0 (−0.5, −2.0)	.001
GH (rest)	1.5 (1.0, 2.0)	1.0 (0.5, 1.5)	.03
GH (strain)	2.0 (1.5, 2.5)	1.5 (1.0, 2.0)	.03
PB (rest)	2.5 (1.5, 3.0)	3.0 (2.0, 3.5)	.17
PB (strain)	2.5 (1.5, 3.0)	3.0 (2.0, 3.5)	.17
TVL	9.5 (8.5, 10.5)	10.0 (9.0, 10.5)	.43

Data presented as mean ± SD or median (IQR) except where noted

^aP-values determined using student’s t-test for comparisons of the means and Mann-Whitney U for POP-Q

^bData presented as n (%)

^cAll POP-Q locations measured in centimeters relative to the hymenal ring at rest, which is defined as 0

Table 2 compares MRI measurements between the young and older groups. Pelvic floor depth was similar between groups. Cervix height and mid-paravaginal support were both lower in the older group with moderate to large effect sizes. Measures of levator area were consistently larger in older women irrespective of the plane in which the measurement was performed—total PICS LH area (oblique axial plane) was 23% larger and mid-sagittal levator area was 50% larger in the older vs young group. Three variables had an extremely large effect size: 1) levator bowl volume at rest (d=1.82), 2) V-U index (d=1.72), and 3) posterior portion of the PICS LH area (d=1.50). For all of these variables, measures were largest in the older group and differences were also statistically significant. Resting levator bowl volume was, on average, 83% larger in the older group. This difference, depicted in Figure 4, is largely due to a posterior enlargement, or deepening, of the levator bowl that was present in the older group. This posterior distension also corresponds to the visible difference in LH shape, which we quantified using the V-U index (Figure 3). Based on this index, the PICS LH area most closely corresponds to a “V” shape in the young group and a “U” shape in the older group. ICC for the V-U index was 0.917 (95% CI 0.791, 0.967). The V-U index was significantly correlated to the following MRI measures: LH at rest (r=0.22, p=.03); PICS LH area: anterior portion (r=−0.49, p=.03), posterior portion (r=0.85, p<.0001), and total area (r=0.60, p=.005); and levator bowl volume at rest (r=0.64, p=.002).

Table 2.

MRI measurements in young and older nulliparous women without prolapse symptoms

MRI Measurementsa	Young (n=12)	Older (n=9)	p-valueb	Effect sizec
Support Relative to PICS Lined, cm
Pelvic Floor Depth	1.9 ± 0.8	1.9 ± 1.1	.95	0.03
Cervix Height	−4.2 ± 1.6	−3.0 ± 1.9	.14	0.68
Mid-Paravaginal Location	−2.5 ± 1.3	−1.2 ± 1.4	.05	0.92
Lengths, cm
Genital Hiatus at Rest	2.7 ± 0.6	3.5 ± 0.6	.007	1.34
Genital Hiatus, Maximal Valsalva	3.1 ± 0.6	3.4 ± 0.8	.39	0.39
Levator Hiatus at Rest	5.3 ± 0.8	6.0 ± 0.8	.049	0.93
Levator Hiatus, Maximal Valsalva	57 ± 12	6.1 ± 1.0	.41	0.37
Areas, cm2
PICS Levator Hiatus Area:
Anterior portion	6.7 ± 2.2	5.6 ± 2.8	.33	0.44
Posterior portion	14.0 ± 2.9	19.9 ± 4.8	.003	1.50
Total area	20.9 ± 2.5	25.8 ± 5.8	.02	1.10
Mid-Sagittal Levator Area	19.5 ± 8.4	29.3 ± 8.8	.02	1.15
V-U Shape Index (V=0, U=1)	0.35 ± 0.18	0.71 ± 0.23	.001	1.72
Levator Bowl Volume, cm3
Rest	59.2 ± 19.3	108.0 ± 34.5	.001	1.82
Maximal Valsalva	115.6 ± 49.0	134.1 ± 43.1	.37	0.41
Difference from Rest to Maximal Valsalva	55.8 ± 50.6	26.1 ± 50.8	.20	0.69
Pubovisceral Muscle Measures
Total Length, cm	3.2 ± 0.6	3.6 ± 0.5	.13	0.70
Cross-Sectional Area, cm2	0.7 ± 0.2	0.8 ± 0.3	.23	0.55
Fat Content, %	5.6 ± 2.3	7.1 ± 2.5	.16	0.07
Vaginal Wall, cm
Vaginal Length (urethral meatus to anterior fornix)	7.3 ±1.5	5.2 ±1.6	.007	1.33
Mid-Vaginal Width	4.2 ±0.8	3.5 ± 0.9	.08	0.82

Data presented as mean ± SD

^aAll MRI measurements made during maximal Valsalva unless otherwise specified

^bP-values determined using student’s t-test

^cEffect size calculated using Cohen’s D, corresponding to small (d=0.20), medium (d=0.50), large (d=0.80), very large (d=1.10), and extremely large (d≥1.40) magnitudes of difference between group means

^dMeasurements made in the vertical plane perpendicular to horizontal reference line (PICS)

Figure 4.

Levator bowl volumes

Other MRI measures with very large effect sizes include GH at rest (d=1.34) and vaginal length (d=1.33). Contrary to the clinical POP-Q measures of GH size, which were larger in the young group, MRI measures of GH at rest and during maximal Valsalva were larger in the older group; for GH at rest, this difference was statistically significant and had a very large effect size. Levator hiatus (LH) lengths at rest and during maximal Valsalva were also larger in the older group; however, these differences were of a smaller magnitude than GH. Vaginal length was 29% shorter in the older group. Only 33.3% (3/9) of the older women, compared to 75% (9/12) of the young women, reported being currently sexually active with vaginal intercourse (p=.09).

Questionnaire results confirmed the absence of bothersome prolapse symptoms in our study cohort. For the POPDI, all women in the young group and seven women in the older group denied experiencing any prolapse symptoms on all six questions. Two women in the older group answered “yes” to needing to push on the vagina or around the rectum to have or complete a bowel movement; this was described as “somewhat” and “moderately” bothersome. For the UDI, eight women in the young group and five women in the older group reported no symptoms. Of those with urinary symptoms, mean bother score was 9.6 ± 4.7 and 11.7 ± 4.1 for the young vs older groups, respectively, with a small effect size (p=.74, d=0.23). Finally, for the CRADI, no colorectal or anal symptoms were reported in seven of the young women and three of the older women. Of those who reported symptoms, mean bother score was 7.3 ± 3.1 and 7.3 ± 3.7 in the young vs older groups, respectively, with an extremely small effect size (p>.99, d=7.8 × 10⁻⁹). Prevalence of constipation based on the Bristol Stool scale was 20% (2/10) in the young group and 0% in the older group (p=.47). Results from the WHOQOL-BREF indicated similar quality-of-life between the young and older women in the domains of physical health (17.7 ± 1.5 vs 17.8 ± 1.3, p=.85, d=0.09) and social relationships (16.5 ± 2.1 vs 17.3 ± 2.1, p=.46, d=0.35). There was a trend toward higher quality-of-life scores among older women in the domains of psychological health (16.1 ± 2.0 vs 17.5 ± 1.2, p=.10, d=0.81) and environment (17.4 ± 2.5 vs 18.5 ± 1.5, p=.27, d=0.53).

Discussion

In this pilot study, we aimed to perform a multi-faceted analysis of age-related functional and structural pelvic floor changes by comparing two groups of nulliparous women. Our results did not support the hypothesis that aging is associated with significantly smaller and weaker levator ani muscles. As expected, handgrip strength was lower among older women; however, it was not correlated with pelvic floor muscle strength. Our hypothesis that genital hiatus would be larger among older women was true only for MRI measures of genital hiatus at rest. The most significant difference we identified was levator bowl volume, with resting levator bowl volume over 80% larger in the older women.

While measures of the genital and levator hiatuses are clearly important and have been an important mainstay of pelvic floor studies, our results suggest that 1D and 2D measures alone capture only a partial view of age-related changes to the pelvic floor. The V-U index is a new parameter that quantifies the 2D change in levator shape with age. The high correlation between the V-U index and resting levator bowl volume demonstrates that these parameters are related. However, as a 3D measure, levator bowl volume provides a more comprehensive measure of levator shape. A few prior studies have shown that levator bowl volume increases with prolapse [26,27]; however, to our knowledge, this is the first study assessing the impact of aging on this measure in the absence of childbirth-related changes. Levator bowl volume at rest was significantly larger in older women, but the 17% larger volume during maximal Valsalva was not significant at our sample size, nor was the increase in volume from rest to maximal Valsalva. This suggests that with aging, the entire pelvic floor descends so that resting volume more closely resembles that observed during maximal Valsalva. Therefore, our results support the “sagging bowl hypothesis,” which supposes that aging leads to a deepening of the resting levator bowl. A possible explanation is that aging may impair the ability of the pelvic floor muscles to maintain an “elevated” position at rest due to a decrease in resting muscle tone. During maximal Valsalva, the levators relax and the pelvic floor is subjected to maximal intraabdominal pressure loading. During this loaded state, we hypothesize that the elastic limit of the connective tissues determines maximal straining levator bowl volume. If the passive tissue properties (i.e. elastic limit of the connective tissues) are similar between groups, straining levator bowl volume would also be similar, as our results show. Further research of pelvic floor connective tissues is needed to test this hypothesis. Our results showing a generalized descent and posterior distension of the pelvic floor is different from prior studies reporting localized eventrations in the levator ani muscles seen in women with pelvic floor dysfunction [28]. Specifically, we noted a posterior distension of the levators in an area cephalad to the pubovisceral and puborectal muscles and more in the region of the iliococcygeus. This is not a phenomenon fully captured by hiatal measurements.

Pubovisceral muscle CSA measurements in our study were not significantly affected by aging, which differs from a prior study by Alperin et al. reporting a 50% reduction with histology [8]. Differences in the ways in which CSA was measured may account for this discrepancy, as Alperin’s study used a calculation to determine PVM CSA in excised cadaveric tissue specimens. They also found a significant increase in intramuscular collagen with aging (51% increase in the iliococcygeus and 72% increase in PVM), which we were unable to assess. However, we were able to measure pelvic floor muscle strength, which would be expected to decline with decreased muscle CSA. Consistent with our results for PVM CSA, we failed to find a significant decrease in Kegel augmentation force among older women. This suggests that the levators and the hiatal closure mechanism may undergo different age-related changes than other skeletal muscles in the body, or that voluntary muscle activation measured with current testing strategies do not fully assess muscle function. Our observation regarding the change in levator hiatus shape from “V” to “U” with aging may reflect a decrease in overall tone or an increase in length in the levators; however, we cannot draw any definitive conclusions based on the current study.

Both aging and parity are key risk factors for prolapse and prior data show that childbirth increases genital hiatus size [29,30]. We know pubovisceral muscle avulsion, which can occur with vaginal delivery, is associated with a 3.5-fold increased prevalence of prolapse and also increases genital hiatus size [31–33]. However, nearly half of women with prolapse do not have significant levator damage, which suggests additional mechanisms other than muscle injury must also play a role. Our data suggest that aging increases the levator bowl volume independent of childbirth changes to the genital hiatus—this finding may provide insight into the age-related contribution to prolapse. Figure 5 shows our proposed conceptual model for pelvic floor changes that result from aging and parity. In the top row, aging, in the absence of childbirth, is associated with increased levator bowl volume but not a significant change in genital hiatus size. Conversely, vaginal parity (left column) is associated with an increase in genital hiatus size that might occur without necessarily changing the posterior aspects of the bowl. In this model, prolapse (bottom right) results from a combination of both enlarged genital hiatus size (childbirth effect) and increased levator bowl volume (aging effect). More studies are needed to test the validity of this model.

Figure 5.

Proposed conceptual model for pelvic floor changes resulting from aging and vaginal parity

Several limitations to our study warrant discussion. As a pilot study, our sample size is small and precludes the use of multivariable logistic regression. We were unable to discern from our data if the increase in resting levator bowl volume is continuous over time or whether it is subject to a threshold effect. Additionally, Kegel augmentation force may not represent the full extent of pelvic floor muscle function, contractility, or the ability of the muscles to maintain tone. We were unable to independently assess the impact of age-related factors such as menopause-related hormonal changes. Our cohort was mostly Caucasian, which limits generalizability. However, it is strengthened by the fact that our groups differed in age by an average of 50 years, which provides for comparison of phenotypic extremes. Furthermore, we performed a variety of measures that included standard clinical evaluation, validated questionnaires, pelvic floor function, and state-of-the-art 3D pelvic floor imaging to provide a comprehensive analysis of pelvic floor structure and function.

In summary, levator bowl volume at rest is strongly affected by age and is over 80% larger in women >70 compared to women <40. Our study suggests that age affects portions of the posterior compartment not fully assessed by conventional clinical measures, nor by 1D or 2D MRI measures. The levator hiatus undergoes a V-to-U transformation with age, and overall, the pelvic floor appears to undergo a generalized posterior distension. This study provides new insight regarding specific age-related changes to pelvic floor support and may help provide a mechanistic explanation of why aging is a risk factor for prolapse.