Mechanisms of Hiatus Failure in Prolapse: A Multifaceted Evaluation

Abstract

INTRODUCTION AND HYPOTHESIS

We investigated whether factors influencing pelvic floor hiatal closure are inter-related or independent, hypothesizing that a) hiatus size is moderately correlated with levator defect, pelvic floor muscle strength, and change in hiatus size with contraction; and b) urogenital hiatus (UGH) and levator hiatus (LH) measures are similar in patients with anterior wall (AW) and posterior wall (PW) prolapse.

METHODS

This cross-sectional case-control study included subjects with AW prolapse (n=50), PW prolapse (n=50), and normal support (n=50). Hiatus measurements and levator defects were assessed on MRI and vaginal closure force was measured with an instrumented speculum. Pearson correlation coefficients and simple and multivariable linear regression models were performed.

RESULTS

During contraction, LH narrowed 47% more in the PW compared to AW group (p=0.001). With straining, LH lengthened 34% more in the PW than AW group (p<0.001). With straining, UGH and LH lengthening that was greater by 72% and 44% in those with Major compared to No/Minor defect (p<0.001 and p=0.004).

Contraction strength explained, at most, 4% of UGH (r=0.17) or LH (r=0.20) shortening during contraction (r=0.17 and r=0.20, respectively), indicating that these factors are largely independent.

After controlling for prolapse size, resting UGH and levator defect status were associated with straining UGH (p<0.001, p=0.004), but muscle strength and resting tone were not.

CONCLUSIONS

Hiatus measures are complex and differ according to prolapse occurrence and type. They are, at best, only weakly correlated with pelvic floor muscle strength and movement during contraction.

Brief Summary:

The interactions between levator ani defects and functional measurements—urogenital and levator hiatus size at rest—during muscle contraction and strain are complex.

INTRODUCTION

Enlargement of the pelvic floor hiatuses has emerged as an important factor in the occurrence of pelvic organ prolapse (POP) [1]. An enlarged hiatus after giving birth has been shown to be associated with prolapse later in life [1–4], and a persistently enlarged hiatus after prolapse surgery is linked with operative failure [5,6]. Therefore, understanding the structural and functional factors associated with hiatal enlargement is critical to understanding the cause of prolapse and operative failure.

Levator ani injury sustained during vaginal delivery is one factor associated with an enlarged hiatus [7–9]; it is also associated with decreased pelvic floor muscle (“muscle”) strength [9–12] and pelvic organ prolapse [13–17]. It has been assumed that an enlarged hiatus explains the connection between levator ani defect (“levator defect”) and prolapse [18]. Evidence is sparse, however, regarding the inter-relationships of a) hiatus size at rest, with contraction, or with strain, b) visible muscle injury, and c) measured muscle contraction force. It is not clear how much muscle injury and functional assessments are related, or whether they are independent factors. Furthermore, there are 2 anatomically distinct hiatus measures that have been identified in the pelvis, the levator and urogenital hiatuses (LH and UGH, Fig. 1). Each has been associated with prolapse but data is lacking regarding the relationship between the two hiatus measures and whether these change proportionally with muscle contraction, strain, and in the presence of prolapse.

Figure 1.

Mid-sagittal MRI measurements of urogenital hiatus and levator hiatus at Rest, Kegel, and Strain

UGH=urogenital hiatus; LH=levator hiatus

These inter-relationships are complex. Although increasing hiatus size has been associated with decreased muscle strength [12], Braekken and colleagues [19] found that muscle function (resting tone and strength) explained only 26% of the variance in levator hiatus (LH) area. Similarly, in a smaller subset of the current study population, though UGH and LH measures were found to be larger with valsalva in subjects with prolapse compared to controls, the presence of a levator defect was found to only explain 12% of the variance in hiatus size [7,20]. This suggests that the relationship between muscle strength, levator defect status, and the hiatus may be less straightforward than previously thought. There is also evidence that hiatus measures differ depending on the type of prolapse present (i.e., anterior- vs. posterior-predominant) [21,22]. There has not, however, been a comprehensive evaluation of LH or UGH size at rest, during strain, and with contraction that takes into consideration prolapse type, muscle strength, and presence of a levator defect.

Our objective was therefore to determine whether several Level III factors influencing hiatal closure (muscle injury, strength and hiatus size) are all related to one or are independent factors. Further, we sought to explore differences between women with anterior wall-predominant (“anterior prolapse”) and posterior wall-predominant prolapse (“posterior prolapse”) compared to asymptomatic controls with normal support. We hypothesized that a) hiatus size is moderately correlated with levator defect, muscle strength, and hiatal narrowing with contraction; and b) hiatus measures are similar in patients with anterior and posterior prolapse.

MATERIALS AND METHODS

This was a cross-sectional case-control study of 50 women with anterior prolapse, 50 with posterior prolapse, and 50 asymptomatic controls. Patients underwent magnetic resonance imaging (MRI) scans and pelvic floor function testing in a single unit as part of institutional review board-approved studies performed at the University of Michigan (HUM0004344). Written informed consent was obtained from patients for the original study. Women were included as cases if they had a uterus and symptomatic prolapse of the anterior or posterior wall descending to at least 1 cm below the hymen on exam. For inclusion in the anterior prolapse group, the posterior wall had to lie above the hymen. Similarly, in the posterior prolapse group, the anterior wall had to lie above the hymen. For inclusion in either case group, the most distal anterior or posterior wall point had to lie below point C in the Pelvic Organ Prolapse Quantification (POP-Q) system. Women were excluded as cases if they had a history of a hysterectomy, prior surgery for pelvic organ prolapse, or performed the wrong maneuver (strained) when asked to perform a pelvic muscle contraction. Women were included as controls if they had no symptoms of prolapse and no prolapse to or beyond the hymen on exam, with point C < −4.

MRI technique has been described previously [23]. In brief, a 3-T Achieva scanner was used to obtain mid-sagittal MRI sequences in the supine position at rest and with prompting for a pelvic muscle contraction and for maximum strain. For the strain sequences, each patient was asked to push to recreate the maximum extent of her prolapse and to hold this for approximately 17 seconds. If a prolapse was not present that matched the clinical POP-Q measures, it was repeated. The dynamic MRI sequences were reviewed, and linear hiatus measures made at the point of maximum contraction and maximum strain (Figure 1). The UGH was measured as the shortest distance from the arcuate pubic ligament at the inferior aspect of the pubic bone to the ventral aspect of the perineal body; the LH was measured as the shortest distance from the arcuate ligament to the ventral surface of the levator ani. These anteroposterior diameter measurements were taken at rest, at maximum strain, and at maximum contraction in the mid-sagittal plane using eFilm®. The change in hiatus length from rest to contraction and strain was also calculated (reported as ‘ΔKegel’ and ‘ΔStrain’ values).

Levator defects were determined by review of proton density axial and coronal MRI images made at 5 mm intervals [24]. Severity of levator defect was graded 0–3 on each side, with a possible total of 0–6. As described previously, 0=no defect; 1=less than half of the muscle missing; 2=more than half of the muscle missing; and 3=total or near total muscle loss on that side [23]. Each subject’s levator defect was categorized by the total score as ‘None’ (0), ‘Minor’ (1–3), or ‘Major’ (4–6).

Patients underwent pelvic floor muscle resting tone and strength assessment with use of an instrumented vaginal speculum [25]. Resting force and Kegel augmentation force (increase in force above resting) were recorded in Newtons (N). Each patient was prompted to perform Kegel contractions three times, and an average was calculated. All patients also underwent complete POP-Q evaluation by a urogynecologist. Demographics, body mass index (BMI), and parity were collected for each subject.

Descriptive statistics stratified by prolapse group and levator defect status were obtained and distributions between groups were compared with one-way analyses of variance or Chi-squared tests. Cohen’s d effect sizes were used to understand group differences. Paired t-tests were used to compare UGH and LH measurements. Pearson correlation coefficients assessed relationships between muscle strength, hiatus measures, levator defect status, and prolapse size (maximum POP-Q point in centimeters) in the entire cohort to represent a full range of support, from normal to significant prolapse. For the purposes of this mechanistic study, absolute value Pearson correlation r values were interpreted as follows: <0.30=no correlation; 0.30–0.49=weak correlation; 0.50–0.69=moderate correlation; and ≥0.70=strong correlation [26]. A series of simple and multivariable linear regression models assessed associations between UGH size at maximum strain and the following predictors: Kegel augmentation strength, resting tone, maximum POP-Q point, UGH at rest, and levator defect status (None/Minor vs. Major). R² values for the simple and multivariable regression models indicated the proportion of variance explained in UGH size at maximum strain by the predictor(s). A priori, a 95% confidence interval (CI) was set and a power calculation indicated that 50 subjects in each group (anterior prolapse, posterior prolapse, and controls) had at least 80% statistical power to detect moderate correlations (r≥0.5) and medium effect sizes (d≥0.57), which are of clinical interest.

RESULTS

The mean age (±SD) for the cohort was 57±11 years. Controls, anterior prolapse, and posterior prolapse subjects were primarily Caucasian and of similar age (Table 1). Those in the anterior prolapse group had one more vaginal birth than in the other groups and those in the posterior prolapse group had a slightly higher BMI. Women in the anterior prolapse group had a prolapse that was approximately 1 cm larger than the posterior prolapse group. A major levator defect was 2.8 times more common among those with prolapse compared to controls (p=0.002), but the difference in frequency of major levator defect between anterior and posterior prolapse groups was small (6%, p=0.54).

Table 1.

Baseline demographics and POP-Q values across prolapse groups

Characteristics	Controls n=50	Anterior Prolapse n=50	Posterior Prolapse n=50	p-value
Age, years	55.2±11.1	57.1±11.1	58.1±10.4	0.40
Race				0.17
Caucasian	43 (86)	43 (86)	48 (96)
Other	7 (14)	7 (14)	2 (4)
Paritya	2 (1–7)	3 (1–7)	2 (1–6)	0.04
Body mass index, kg/m2	26.9 (5.2)	26.5 (4.6)	28.9 (5.2)	0.04
POP-Q points, cm
Ba	−1.8±0.7	2.9±1.9	−1.1±1.5	<0.001
C	−6.8±1.4	−1.5±4.1	−4.8±2.6	<0.001
D	−8.9±1.8	−5.8±2.0	−6.2±3.2	<0.001
Bp	−1.9±0.7	−1.1±1.4	2.1±1.5	<0.001
Maximum prolapse	−1.6±0.8	3.0±1.9	2.1±1.5	<0.001
Major levator defect	7 (14)	21 (42)	18 (36)	0.006

Data presented as Mean±SD or n (%) except where noted

^aMedian (range)

POP-Q=Pelvic Organ Prolapse Quantification

Hiatus and Prolapse

Resting UGH in the anterior prolapse group (Figure 2A) was 21% larger than in the posterior prolapse group (35.5 vs 29.4 mm, p<0.001), whose size was similar to controls (p=0.39). Resting LH was 8% larger in the women with prolapse (55.9 and 55.8 mm in AW and PW groups vs 51.6 mm in controls, p=0.007), but similar between anterior and posterior prolapse groups (p=0.97).

Figure 2. Difference and change in urogenital hiatus and levator hiatus with Kegel and Strain from rest by prolapse status.

A) Length of hiatus at rest. B) Change in length of hiatus from rest with Kegel and Strain. Kegel=above 0 axis; Strain=below 0 axis. Standard error bars shown.

UGH=urogenital hiatus; LH=levator hiatus; PW=posterior wall prolapse; AW=anterior wall prolapse

With contraction, UGH narrowing was similar in women with and without prolapse (p=0.36), but the LH narrowed 47% more in women with posterior prolapse compared to anterior prolapse (9.1 vs 6.2 mm, p=0.001) (Figure 2B). During straining, the UGH in women with prolapse increased 2.3 times (230%) more than controls (p<0.001) and was similar in anterior and posterior prolapse groups (p=0.84). With strain, the LH lengthened 2.1 times (210%) more in the posterior prolapse group compared to controls and 34% more in the posterior than the anterior prolapse group (p=0.02) (Figure 2B).

Pelvic floor closure force measurement (Figure 3) revealed no difference in resting tone between prolapse groups, but controls had 53% greater Kegel augmentation strength compared to those with prolapse (p=0.008). Kegel augmentation strength did not substantively differ between anterior (1.3N) and posterior prolapse (1.8N) groups (p=0.09).

Figure 3.

Pelvic floor muscle resting tone and Kegel augmentation by prolapse type (A) and levator defect status (B)

PW=posterior wall prolapse; AW=anterior wall prolapse. Standard error bars shown.

Levator Defect Status

Those with a Major levator defect had a 13% larger resting UGH (by 3.9 mm) and 13% larger resting LH (by 6.7 mm) than those with No/Minor levator defect (p=0.006 and p<0.001, respectively) (Figure 4). The amount of UGH or LH shortening with contraction did not differ by defect status, but UGH and LH lengthening was greater by 72% and 44% respectively among those with a Major compared to No/Minor levator defect (p<0.001 and p=0.004).

Figure 4. Levator defect status and hiatal dimensions.

A) Length of hiatus at rest. B) Change in length of hiatus from rest with Kegel and Strain. Kegel=above 0 axis; Strain=below 0 axis. Standard error bars shown.

UGH=urogenital hiatus; LH=levator hiatus

Resting vaginal closure force did not differ between levator defect groups (Figure 3). However, those with No levator defect had twice the Kegel augmentation strength compared to those with a Major levator defect (2.5 vs 1.3 N, p<0.001, effect size 0.75). Kegel augmentation strength was similar in those with a Minor and Major levator defect (p=0.32).

Relationships between Hiatus Measures, Muscle Function, and Levator Defect

The Pearson correlation matrix (Figure 5) depicts the inter-relationships between these different factors, evaluating strength of association between static and dynamic hiatus measurements; pelvic floor muscle resting tone and strength; and levator defect score (measured from 0– 6) for the overall cohort, including patients with and without prolapse. Of the 36 comparisons, there were three with moderate correlations; these were the positive correlations between LH and UGH resting values, LH and UGH with strain, and LH and UGH with contraction (r=0.54–0.64). Weak correlations were observed for three comparisons: positive correlations between levator defect score and LH rest (r=0.36) and between levator defect score and ΔUGH strain (r=0.33) and a negative correlation between LH rest and Kegel strength (r=−0.35). Otherwise, there was no significant correlation between voluntary muscle contraction strength and any other hiatus measures, including ΔKegel for UGH or LH (r=0.17 and 0.20). Similarly, levator defect score was not correlated with strength or other hiatus measures aside from weak correlations with ΔUGH strain and LH rest as described above (r ranged from −0.27 to −0.28).

Figure 5.

Pearson correlation coefficients (r) comparing selected MRI measures, strength measures, and levator defect status

UGH=urogenital hiatus; LH=levator hiatus; LA=levator ani; levator defect=levator ani defect; PFM=pelvic floor muscle

Contributors to UGH at Maximum Strain

Considering all measures of hiatus size, muscle strength, levator defect status, age, BMI, and parity, we found that the variable most strongly associated with prolapse size as assessed by the maximum POP-Q point was UGH at maximum strain that explained 38% of variation (r²=0.38). The amount of UGH lengthening with straining (ΔUGH strain) was also moderately correlated with prolapse size (r²=0.27), but all other hiatus measures had weak or no correlations with prolapse size. Pelvic floor muscle (PFM) closure force and Kegel augmentation strength were not correlated with prolapse size (r²=0.004 and, r²=0.06, respectively).

To determine predictors of UGH at maximum strain, associations with UGH strain were assessed with correlations and simple linear regression models. There was a strong correlation between UGH strain and ΔUGH strain (r=0.84) and moderate correlations with UGH rest (r=0.68), UGH Kegel (r=0.66), and LH strain (r=0.65). There were weak correlations with LH rest, LH Kegel, ΔLH strain, and levator defect status and no correlation with PFM closure force or Kegel augmentation strength (r=0.09 and r=−0.18, respectively).

A multivariable linear regression model (Appendix 1) investigated the degree to which the strength and muscle defect factors were related to UGH size at maximum strain. When controlling for prolapse size, resting UGH and levator defect status were associated with straining UGH, but Kegel strength and resting tone were not. For every 1 cm increase in maximum POP-Q point, there was a 1.7 mm increase in UGH at maximum strain. Those with a Major levator defect had UGH strain measurements that were on average 5 mm larger than those with No/Minor levator defect. This combination of variables explains 61% of the variance in UGH size at maximum strain (adjusted r²=0.61). A series of simple linear regression models were used to obtain individual r² values for each measure as follows: UGH rest=0.46; maximum POP-Q point=0.38; levator defect status=0.15; Kegel strength=0.03, and Resting tone=0.007.

DISCUSSION

This study demonstrates that the relationship between urogenital and levator hiatus size, pelvic floor muscle strength, and levator defect is complex. These measures are largely independent of one another and differ according to prolapse occurrence and type. We reject our hypothesis that hiatus size is moderately correlated with levator defect, muscle strength, or hiatal narrowing with contraction. Hiatus measures were only weakly correlated with muscle strength, and when controlling for prolapse size, levator defect and muscle strength explained only 15% and 3% of the variance in hiatus size, respectively. We also found that there was greater shortening and lengthening of the levator hiatus among women with posterior-predominant compared to anterior-predominant prolapse, but pelvic floor muscle strength was comparable. Overall, the concept that muscle strength and defect status are the strong determinants of hiatal size is not supported.

The largest correlations—those between the LH and UGH—are not surprising, because they both reflect changes in the pelvic floor aperture. In this group of women with and without prolapse, PFM strength was at best only weakly correlated with hiatus size. Furthermore, when controlling for prolapse size, levator defect status and PFM strength explained only 15% and 3% of the variance in hiatus size, respectively. We also found that there were differences in hiatus measures, but comparable strength between those with anterior and posterior prolapse.

While it is logical that PFM injury would be associated with reduced strength, which then causes a dilated hiatus, this paradigm may be overly simplistic. The current study shows that PFM function and levator defect explain only a small percentage of the variance in hiatus size, supporting the findings of Braekken [19] and Nandikanti [7].

Our findings also show that hiatus size and muscle strength are independent factors rather than two similar measures of a single phenomenon. This supports Handa’s recent report that LH area and PFM strength on perineometer independently relate to the presence of prolapse [18]. Handa’s study also found that 61% (95% CI, 34%–106%) of the association between levator defect and prolapse was mediated by LH area and PFM strength; therefore, there might still be other factors at play, such as those possibly related to denervation of the LA muscle [27].

When controlling for the effect of prolapse size in our model, we were not surprised to find that resting UGH and levator defect were associated with UGH strain. It was more striking that differences in muscle strength and resting tone were not meaningful, demonstrating that these measures vary independently from hiatus measures.

The differences in hiatus function between anterior and posterior prolapse are also interesting. Women with anterior prolapse have a larger UGH but not LH at rest, while those with posterior prolapse have more total excursion (both shortening and lengthening) of the LH. These are simply two different assessments of a complex shape—the levator bowl. Further exploration of these phenomena using newer shape analysis techniques is needed [28].

The 25% anatomical failure rate with the gold standard operation for prolapse[29] indicates that there are factors not being addressed by our current surgical approaches that limit our ability to achieve normal support. There has been extensive work on addressing Level I apical support, but less emphasis on understanding factors related to hiatal closure. If, as evidence indicates [5], a persistently enlarged hiatus is associated with operative failure, then understanding the factors involved in hiatal closure is of primary importance in looking for ways to improve treatment outcome and prevention.

These findings suggest there are factors at play that determine hiatus size and function other than PFM strength and the presence of an identifiable levator defect on imaging. These may include the role of connective tissue; the presence of a dilating prolapse; alterations in the perineal membrane/perineal body; or other muscle factors such as denervation, perfusion, and individual differences in the bony pelvis—all represent important areas for future investigation. An overall theoretical model for the complex interactions among these factors is needed.

Strengths of our study are its multifaceted evaluation, including static and resting measures of both the UGH and LH in high-resolution MRI scans with close supervision by study personnel to assure proper acquisition, in addition to its well-established evaluation of levator defect, strength assessment with an instrumented speculum, and physical examination data. We not only compared hiatus function across the cohort, but also according to prolapse type, which has not been explored to this extent in the past. Greater total excursion of the LH among those with posterior prolapse suggests a mechanism unique to this type of prolapse that should be explored further in future studies.

Our findings are limited by inherent difficulties in capturing pelvic floor muscle function. By analogy, knowing grip strength and hand size would not predict piano performance, and similarly, currently available techniques do not evaluate all aspects of pelvic floor function. Similarly, there is no perfect way to measure the complex three dimensional hiatuses in the pelvis. We chose to compare linear mid-sagittal MRI hiatus measurements, but transverse hiatus diameter may also play a minor role. [add ref #1 again here DeLancey 1998] It is also worth noting that women who performed the wrong maneuver (e.g., strain when asked to perform a Kegel) were excluded from the study. This may have biased our sample by excluding some patients who were unable to activate the levator muscle because of severe muscle or nerve damage. It is also important to note that in our cohort, as in the general population [30], those with anterior type prolapse were more likely to have a component of apical prolapse compared to those with posterior prolapse which may further be related to differences seen in hiatus measures and strength between these groups (Table 1). We also do not have information about participants’ prior experience with pelvic floor muscle training which likely affects hiatal narrowing [31]. Finally, the fact that our study included specific groups and was not population-based must be kept in mind when interpreting our results.

Pelvic floor muscle strength, hiatal narrowing with muscle contraction, and levator defect status were each related to prolapse, but appear to be independent factors that were only weakly correlated with hiatus size. Further studies are needed to investigate the complex interplay between these factors, hiatus function, and prolapse.

Appendix 1.

Multivariable linear regression model for UGH size at maximal strain

Variable	Coefficients (Beta)	p-valuea	95% CI
Maximum POP-Q point (cm)	1.72	<0.001	1.02–2.41
Kegel strength (N)	0.38	0.41	−.52–1.27
Resting tone (N)	0.16	0.80	−1.04–1.36
UGH rest (mm)	0.87	<0.001	0.67–1.06
Major levator defectb	5.07	0.004	1.66–8.49

^aR² for the model: 0.63 (adjusted R²: 0.61)

^bReference: No/Minor levator defect

POP-Q= Pelvic Organ Prolapse Quantification; UGH=urogenital hiatus; levator defect=levator ani defect