- Email: [email protected]
Size of the urogenital hiatus in the levator ani muscles in normal women and women with pelvic organ prolapse
Size of the Urogenital Hiatus in the Levator Ani Muscles in Normal Women and Women With Pelvic Organ Prolapse JOHN 0. L. DELANCEY,
MD, AND WILLIAM
Objective: Compare urogenital hiatus size in normal women and women with pelvic organ prolapse. Methods: The sagittal and transverse urogenital hiatus diameters were measured and hiatus area calculated in 300 women whose support was scored using a modified Baden system. Results: In women with normal support and without prior surgery, hiatus area was 5.4 cm* (k1.71 standard deviation [SD], IZ = 197). In women with uncorrected clinical prolapse (grade 2-31, the area of the hiatus was enlarged (9.6 cm2 -C 3.97, n = 34, P < .05) and became larger with progressive prolapse (grade 0,5.4 cm2 f 1.71, n = 197; grade 1,7.3 cm2 f 1.91, IZ = 27; grade 2,8.3 cm2 f 2.45, n = 18; grade 3,11.0 cm* +: 4.90, n = 16). When matched for age and parity, prolapse patients had a larger hiatus than normal women. Women with recurrent prolapse had a larger hiatus than cured women (13.3 cm2 f: 3.86 n = 8 compared with 8.1 cm* f 2.44 II = 16 P C .05) or women with recurrence after one operation (8.9 cm2 f 1.77 iz = 18 P < .05). Hiatus size in patients surgically cured (8.3 cm* f 2.44, n = 16) did not return to normal (5.4 cm2 f 1.71, n = 197, P C .05). Increasing area of the hiatus was correlated with an increase in anteriorposterior diameter (? = 0.9, P c .05), was less attributable to increased transverse diameter (? = 0.6, P < .05), and was not related to thickness of the perineal body (2 = 0.0, P > .05). Conclusion: Increasing pelvic organ prolapse is associated with increasing urogenital hiatus size; the hiatus is larger after several failed operations than after successful surgery 0 1998 by or single failure. (Obstet Gynecol1998;91:364-8. The American College of Obstetricians and Gynecologists.)
Damage to the levator ani muscles is associated with pelvic organ prolapse, as has been documented on dissection,i with radiography,’ and in neuromuscular studies.3 Because pelvic organ prolapse occurs through the urogenital hiatus in the levator ani muscles, it seems From the Division of Gynecology, Department of Obstetrics and Gynecology, University of Michigan Medical School, Ann Arbor, Mickigun. Supported by National Institutes of Health grant ROl DK 47516.
364
0029.7844/98/$19.00 PI1 50029-7844(97)00682-O
WV.HURD, MD possible that loss of the levator ani muscles’ ability to close the hiatus is a factor contributing to prolapse. The boundaries of the urogenital hiatus within the levator ani muscles are easily palpable on pelvic examination (anteriorly, the pubic bones; laterally, the medial margins of the levator ani muscles; and dorsally, the perineal body). It is possible, therefore, to measure directly the size of the urogenital hiatus on pelvic examination. This study was undertaken to determine the size of the urogenital hiatus in women with normal support and to quantify the degree of its enlargement in women with prolapse.
Materials and Methods A total of 300 women 18-93 years of age (mean 44.7, 2 15.0 [standard deviation, SD]) parity O-7 (1.7, 1-1.5) examined at one of our outpatient practices (JOLD) during a 12-month-period gave consent to participate in this study. Of these women, 286 classified themselves as white, 9 as black, 3 as Hispanic, and 2 as Asian. Included were patients with primary and recurrent prolapse as well as patients presenting for annual examination and problems not related to prolapse. Support of the pelvic organs was determined specifically during pelvic examination and was used to classify patients for study purposes. The number of previous surgical repairs for prolapse was noted. The status of each woman’s pelvic support was determined by observing the location of the cervix or the vaginal apex, anterior, and posterior vaginal walls, with the patient straining effectively enough to develop the prolapse to its maximum extent. The posterior blade of a Graves’ speculum was used to examine separately the anterior and posterior walls as well as to determine the location of the uterine cervix during Valsalva maneuver. Women who could not generate sufficient pres-
Obstetrics b Gynecology
Table
1. Grading
System
Grade 0 1 2 3
Description No descent beyond the midvaginal line Descent past the midvaginal line but not beyond the plane of the hymenal ring Descent through but not more than 4 cm below the plane of the hymenal ring Descent more than 4 cm below the ring
to reveal the full extent of their prolapse or to stress fully the normal supports in the lithotomy position were examined while standing. The degree of prolapse present was quantified by measuring the distance between the most dependent portion of the prolapse and the hymenal ring. Definitions are listed in Table 1 and are based on a modification of the system of Baden and Walker.4 For purposes of classification, grade 0 was considered normal and grades 2 and 3, definite clinical prolapse. Grade 1 was consistent with the laxity common in asymptomatic parous women. The size of the urogenital hiatus in the levator ani muscles was measured with a ruler to the nearest 0.5 cm by palpating the margins of the hiatus (Figure 1). The anterior-posterior diameter corresponded to the distance from the pubic symphysis to the palpable margin of the firm, dense connective tissue of the perineal body (not the perineal skin). The distance between the medial borders of the levator ani muscles was the transverse diameter. The area of the hiatus was estimated by using these diameters to calculate the area of an oval. In addition, the thickness of the perineal body was measured as the dimension from the ventral border of the perineal body to the anal verge. In individuals with prolapse, the protruding prolapse was replaced into the pelvis, and the acute dilation of the introitus allowed to recover before making measurements. During the data acquisition, 28 patients were examined on two occasions separated by an average of 4 months, during which time they were not operated on and their pelvic support did not change. During the second examination, the previous measurements were not available, and test-retest reliability of these measurements had acceptable repeatability with Y’ = 0.81, P < .05. Only first measurements were used in analyzing the relationship between prolapse and hiatus size so that patients would not be represented twice. Statistical analysis was carried out using Statview statistical software (Abacus Concepts, Berkeley, CA). Analysis of group differences was carried out for multiple comparisons by using analysis of variance, with post hoc analysis performed using Fisher protected least significant difference test. The relationship between hiatus diameters and hiatus area was assessed by
sure
VOL.
91, NO.
3, MARCH
1998
calculating the correlation coefficient between these measures. To analyze the effect of age on prolapse size, 34 patients not previously operated on (cases) were matched with an equal number of women with normal support and no previous prolapse surgery. Matching was carried out for age (within 5 years) and vaginal parity. Matching resulted in groups of similar age (controls 53 ? 13.3 years compared with cases 55 2 12.4 years, P > .05) and parity (2.3 ? 1.5 compared with 2.9 2 1.6, P > .05). Paired t test was used to assess the significance of the difference between these matched groups.
Results Support of the pelvic organs, group-specific demographics, and previous prolapse surgery history are shown in Table 2. The area of the urogenital hiatus in the levator ani among women not previously operated on is shown in Figure 2. Of the patients who had significant prolapse (grades 2 and 3), only 26% fell within the normal range (mean ? 2 SD; ~8.4 cm’), and only 6% were at or below the mean of 5.4 cm’. The case
Figure 1. Dimensions measured of the urogenital ani muscles. Al’ = anterior-posterior diameter; diameter; PB = perineal body.
DeLancey
and
Hurd
hiatus Trans.
in the levator = transverse
Urogenital Hiatus and Prolapse
365
Table
2. Support
Status
and
Previous
Surgical
No prior Grade
n
prolapse
Age
0
197
39
1
27
46
2 3
18 16
52 61
Data are presented
as mean
k standard
Treatment
for
Prolapse
surgery
Prior
(y)
n
Parity
t 11.8 + 8.8 -c 13.4 -c 16.5
1.3 2.2 2.3 3.3
2 1.2 -c 1.5 ? 1.75 2 1.7
prolapse
surgery Parity
Age (y)
6
56 t 63 5 56 ir 64 2
10 12 14
12.8 16.1
3.2
10.8 16.1
2.4 3.5
2.1
k 2 t 2
1.5 1.4 1.2 1.5
deviation.
control analysis carried out to assess the possibility that age or parity was responsible for difference in hiatus size showed a persistent difference between cases with unrepaired prolapse (mean hiatus area 9.5 cm’ 2 3.8) compared with normal support controls (mean area 5.7 cm’ i 1.4, P < .05). The relationship between prior surgery and prolapse, and size of the hiatus is shown in Figure 3. Among patients with recurrent prolapse, those who had two or more previously failed attempts to correct prolapse had a larger hiatus than those in whom surgery had been successful (P < .05) or those who had only one previous failed repair (P < .05). Even in patients whose prolapse had been cured surgically, the hiatus did not return to normal size (P < .05). The increase in hiatus size was associated more strongly with the changes in the anterior-posterior diameter (? = .9, P < .05) whether measured to the perineal body or the anus (Figure 4). There also was .some increase in transverse diameter as the area enlarged (Y* = .6, P < .05) (Figure 5). There was no correlation between the thickness of the perineal body and the size of the urogenital hiatus (r2 = .0002, P > .05) (Figure 6).
Discussion This study revealed that the size of the urogenital hiatus is larger in women with prolapse than it is in women with normal support. This difference persisted even in those women who have been cured of their prolapse. The enlarged hiatus, therefore, is not entirely a reflection of the presence of the prolapse. Furthermore, the hiatus is progressively larger in women with increasingly severe prolapse and is largest in those individuals in whom several surgical repairs have failed to correct the prolapse. The degree of prolapse present, rather than a women’s presenting complaint, was used in this report because the study’s intent was to correlate changes in size of the hiatus with differing degrees of loss of support. In each woman, the type of prolapse was noted (eg, enterocele and cystocele). Because of the great diversity of findings, this information was not used for analysis, because each group was too small to yield reliable information.
16
16
r
T
L
2
t-
0
0 1
ti
0
1
II=197
n=27
283 n=18
n=16
and
Hurd
Urogenital
-
Failed 1 Oper. n=16
Cured n=16
i
Failed > 1 Oper n=8
Surgical History
Figure 2. Hiatus size in women not operated on previously for prolapse, by grade with standard deviation shown. All clinical prolapses (grades 2 and 3) are summarized at right. *P i .05 compared with normal.
DeLancey
Unrepaired n=34
II=34
Grade
366
Normal n=198
Hiatus
and Prolapse
Figure 3. Influence of prior surgery and its success on hiatus area, with standard deviation shown. Oper. = operation. *P < .05 compared with normal. **Compared with unrepaired, cured, and after one failed repair.
Obstetrics
& Gynecology
8
y = 8.9E-3x
+ 2.3,
r2 = 1.9E-3
7
Perineal Body Thickness
--Y.-+-,
2.5
0
5
7.5
hlarus
__I-
12.5
15
Size
(cm.
sq.)
17.5
20
22.5
.Q
2.5
5
7.5
Hiatus
10
Area
12.5
15
17.5
20
2
(cm. sq.)
4.
Although the firtdilkgs ut this study established an aasociatlon between Moe of the hiatus and pelvic organ prulapse, they did not establish causation, determine the structure that is responsible fur hiatus closure, or uldicate why closure should influence pelvic organ support. Kelly’ observed that it is the levatur ani and Ilut the pttli~~al body ur bulbucdvernous muscle that &ses the vagmdl outlet, and our anatomic observations were corkstent with this observation. The direction of
gr--,-.---’
. ’ N-----t Aws
’
/
/
/. / 6-
rtarlsv~f5~ Diameter (cm.1
,
/
/
/
/
/
,
F-b / / t
1 /-
t
1 /
0 1 .0
I -.--.. 2.5 5
iv 7.5
Hiatus ~~gurr 5. hxieidtlul~ hi&us and the
YI,
NO.
brmern
vwa.tw 10 12.5
Size
15
17.5
20
22.5
(cm. sq.)
the t1d11svelse didmeter of the urogeahiatus. [
VUL
!
v_-
10
~u~~elat~u~l between the sqgttdl dlanneters of the urogenital h~us mrusured frum the pubic borle to either the perineal body (I’S, +J)L,I LL&SJ or the uus (oyei~ burrs) and the ared of the hiatus. (All pktts x+zh sm&u medsureinents are represented by d s~rrgle dot.)
kq+z~e
6
Dora of the
5, MAKLH
1998
Figure 6. Correlation between the thickness of the perineal body the area of the hiatus. (All patients with the same measurements represented by a single mark.)
and are
the muscle fibers within the levator ani muscle are such that muscular relaxation would result in an increase in the distance between the pubic bone and the perineal body and anus. Furthermore, the constant contraction of the levator ani muscle’ is well suited to maintaining continuous closure of the pelvic floor and is similar to the continuous tone in the external anal sphincter. Our data indicating increased anterior-posterior diameters but no change in perineal body thickness further supported Kelly’s concept. The hypothesis that damage to the levator ani muscle plays a role in causing pelvic organ prolapse has been proposed for many years. Direct dissections of these muscles in cadavers with prolapse’ and radiographic studies2 have demonstrated this role. More recently, electromyography has identified partial denervation of the levator ani muscles as being associated with prolapse3 and as possibly explaining the loss of function noted in the earlier studies. Halban and Tandler’s direct anatomic studies’ of the pelvic floor in women with prolapse concluded that injury to the pelvic floor with opening of the so-called pelvic valve was instrumental to developing prolapse. Berglas and Rubin observed that the levator plate loses its horizontal position in patients in association with enlargement of the levator hiatus, and Nichols et al7 pointed out the clinical implications of this change. Because changes in the closure of the urogenital hiatus would be expected to be associated with changes in the inclination of the levator plate, these two changes probably are reflections of the same phenomenon. The finding that an enlarged hiatus is associated with
DeLancey alld Hurd
Ur.ogmitd
Hlutus
ad
Prolapse
367
genital prolapse suggested the following hypothetical mechanism by which damage to the levator ani muscle and enlargement of the urogenital hiatus might influence pelvic support. In the normal individual, the 5-cm* area of the hiatus is filled with the vaginal walls, endopelvic fascia, and urethra. In effect, these organs fill the opening in the muscles, forming a seal in its opening. As the size of the opening within the muscle enlarges, an open space in the pelvic floor at the vaginal opening develops. Because the vaginal canal does not contain air, something must fill this space. This may be a cystocele, rectocele, or the uterine cervix. Once the pelvic outlet has opened and the vaginal wall or cervix lies unsupported at the boundary between abdominal and atmospheric pressure, the endopelvic fascia is called on to attempt to keep the unsupported pelvic organs in place, unsupported by the levator ani muscles. The constant load of abdominal pressure on this connective tissue over time could be expected to cause failure of the connective tissue in certain predisposed individuals, with the development of significant prolapse.8 It should be noted that it is not necessary for the levator ani muscles to be damaged for prolapse to occur, because a few women in our study have overt prolapse despite a normal hiatus. It is possible for the cervix or vaginal wall to prolapse downward because of defective connective tissue supports without damage to the muscles. In this situation, the prolapse simply dilates an otherwise normal pelvic floor by driving a wedge through the urogenital hiatus. This explains the overlap in hiatus size between normal women and some women with prolapse. Nevertheless, the vast majority of women with prolapse have a hiatus that falls outside the normal range, indicating that prolapse in women with a normal hiatus is the exception rather than the rule. Important questions remain to be answered in determining the etiology of pelvic organ prolapse. The factor responsible for enlargement of the urogenital hiatus remains to be determined: Is it neurologic injury to the levator ani muscles, or is it direct muscle trauma as might occur at the time of vaginal birth? What is the
368
DeLancey
and
Hurd
Urogenital
Hiatus
and Prolapse
relationship between abnormally weak connective tissue and these changes in hiatal size? How do pelvic muscle exercise and surgical intervention influence the size of the urogenital hiatus? Now that a simple technique is available to assess urogenital hiatus size, it should be feasible to begin to gather data concerning these issues.
References 1. Halban J, Tandler J. Anatomie und atiologie der genital prolapse beim weibe. Vienna and Leipzig: Wilhelm Braunmuller, 1907. 2. Berglas B, Rubin IC. Study of the supportive structures of the uterus by levator myography. Surg Gynecol Obstet 1953;97:677-92. 3. Smith ARB, Hosker GL, Warrell DW. The role of partial denervation of the pelvic floor in the aetiology of genitourinary prolapse and stress incontinence of urine: A neurophysiological study. Br J Obstet Gynaecol 1989;96:24-8. 4. Baden WF, Walker T. Fundamentals, symptoms, and classification. In: Baden WF, Walker T, eds. Surgical repair of vaginal defects. Philadelphia: JB Lippincott, 1992. 5. Kelly HA. Gynecology. New York: Appleton and Company, 1928: 265-7. 6. Parks AG, Porter NH, Melzak J. Experimental study of the reflex mechanism controlling muscles of the pelvic floor. Dis Colon Rectum 1962;5:407-14. 7. Nichols DH, Milley I’S, Randall CL. Significance of restoration of normal vaginal depth and axis. Obstet Gynecol 1970;36:251-6. 8. Paramore RH. The supports-in-chief of the female pelvic viscera. J Obstet Gynecol Br Emp 1907;13:391-409.
Reprints are not available. Address correspondence to: John 0. L. DeLancey, MD Department of Obstetrics and Gynecology University of Michigan Medical School L4100 Women’s Hospital 1500 East Medical Center Drive Ann Arbor, MI 48109-0216 E-mail: [email protected]
Received Received Accepted
May 22, 1997. in revised form September September 10,1997.
Copyright 0 Gynecologists.
1998 by The American Published by Elsevier
9, 1997.
College of Obstetricians Science Inc.
Obstetrics
b Gynecology
and
MD, AND WILLIAM
Objective: Compare urogenital hiatus size in normal women and women with pelvic organ prolapse. Methods: The sagittal and transverse urogenital hiatus diameters were measured and hiatus area calculated in 300 women whose support was scored using a modified Baden system. Results: In women with normal support and without prior surgery, hiatus area was 5.4 cm* (k1.71 standard deviation [SD], IZ = 197). In women with uncorrected clinical prolapse (grade 2-31, the area of the hiatus was enlarged (9.6 cm2 -C 3.97, n = 34, P < .05) and became larger with progressive prolapse (grade 0,5.4 cm2 f 1.71, n = 197; grade 1,7.3 cm2 f 1.91, IZ = 27; grade 2,8.3 cm2 f 2.45, n = 18; grade 3,11.0 cm* +: 4.90, n = 16). When matched for age and parity, prolapse patients had a larger hiatus than normal women. Women with recurrent prolapse had a larger hiatus than cured women (13.3 cm2 f: 3.86 n = 8 compared with 8.1 cm* f 2.44 II = 16 P C .05) or women with recurrence after one operation (8.9 cm2 f 1.77 iz = 18 P < .05). Hiatus size in patients surgically cured (8.3 cm* f 2.44, n = 16) did not return to normal (5.4 cm2 f 1.71, n = 197, P C .05). Increasing area of the hiatus was correlated with an increase in anteriorposterior diameter (? = 0.9, P c .05), was less attributable to increased transverse diameter (? = 0.6, P < .05), and was not related to thickness of the perineal body (2 = 0.0, P > .05). Conclusion: Increasing pelvic organ prolapse is associated with increasing urogenital hiatus size; the hiatus is larger after several failed operations than after successful surgery 0 1998 by or single failure. (Obstet Gynecol1998;91:364-8. The American College of Obstetricians and Gynecologists.)
Damage to the levator ani muscles is associated with pelvic organ prolapse, as has been documented on dissection,i with radiography,’ and in neuromuscular studies.3 Because pelvic organ prolapse occurs through the urogenital hiatus in the levator ani muscles, it seems From the Division of Gynecology, Department of Obstetrics and Gynecology, University of Michigan Medical School, Ann Arbor, Mickigun. Supported by National Institutes of Health grant ROl DK 47516.
364
0029.7844/98/$19.00 PI1 50029-7844(97)00682-O
WV.HURD, MD possible that loss of the levator ani muscles’ ability to close the hiatus is a factor contributing to prolapse. The boundaries of the urogenital hiatus within the levator ani muscles are easily palpable on pelvic examination (anteriorly, the pubic bones; laterally, the medial margins of the levator ani muscles; and dorsally, the perineal body). It is possible, therefore, to measure directly the size of the urogenital hiatus on pelvic examination. This study was undertaken to determine the size of the urogenital hiatus in women with normal support and to quantify the degree of its enlargement in women with prolapse.
Materials and Methods A total of 300 women 18-93 years of age (mean 44.7, 2 15.0 [standard deviation, SD]) parity O-7 (1.7, 1-1.5) examined at one of our outpatient practices (JOLD) during a 12-month-period gave consent to participate in this study. Of these women, 286 classified themselves as white, 9 as black, 3 as Hispanic, and 2 as Asian. Included were patients with primary and recurrent prolapse as well as patients presenting for annual examination and problems not related to prolapse. Support of the pelvic organs was determined specifically during pelvic examination and was used to classify patients for study purposes. The number of previous surgical repairs for prolapse was noted. The status of each woman’s pelvic support was determined by observing the location of the cervix or the vaginal apex, anterior, and posterior vaginal walls, with the patient straining effectively enough to develop the prolapse to its maximum extent. The posterior blade of a Graves’ speculum was used to examine separately the anterior and posterior walls as well as to determine the location of the uterine cervix during Valsalva maneuver. Women who could not generate sufficient pres-
Obstetrics b Gynecology
Table
1. Grading
System
Grade 0 1 2 3
Description No descent beyond the midvaginal line Descent past the midvaginal line but not beyond the plane of the hymenal ring Descent through but not more than 4 cm below the plane of the hymenal ring Descent more than 4 cm below the ring
to reveal the full extent of their prolapse or to stress fully the normal supports in the lithotomy position were examined while standing. The degree of prolapse present was quantified by measuring the distance between the most dependent portion of the prolapse and the hymenal ring. Definitions are listed in Table 1 and are based on a modification of the system of Baden and Walker.4 For purposes of classification, grade 0 was considered normal and grades 2 and 3, definite clinical prolapse. Grade 1 was consistent with the laxity common in asymptomatic parous women. The size of the urogenital hiatus in the levator ani muscles was measured with a ruler to the nearest 0.5 cm by palpating the margins of the hiatus (Figure 1). The anterior-posterior diameter corresponded to the distance from the pubic symphysis to the palpable margin of the firm, dense connective tissue of the perineal body (not the perineal skin). The distance between the medial borders of the levator ani muscles was the transverse diameter. The area of the hiatus was estimated by using these diameters to calculate the area of an oval. In addition, the thickness of the perineal body was measured as the dimension from the ventral border of the perineal body to the anal verge. In individuals with prolapse, the protruding prolapse was replaced into the pelvis, and the acute dilation of the introitus allowed to recover before making measurements. During the data acquisition, 28 patients were examined on two occasions separated by an average of 4 months, during which time they were not operated on and their pelvic support did not change. During the second examination, the previous measurements were not available, and test-retest reliability of these measurements had acceptable repeatability with Y’ = 0.81, P < .05. Only first measurements were used in analyzing the relationship between prolapse and hiatus size so that patients would not be represented twice. Statistical analysis was carried out using Statview statistical software (Abacus Concepts, Berkeley, CA). Analysis of group differences was carried out for multiple comparisons by using analysis of variance, with post hoc analysis performed using Fisher protected least significant difference test. The relationship between hiatus diameters and hiatus area was assessed by
sure
VOL.
91, NO.
3, MARCH
1998
calculating the correlation coefficient between these measures. To analyze the effect of age on prolapse size, 34 patients not previously operated on (cases) were matched with an equal number of women with normal support and no previous prolapse surgery. Matching was carried out for age (within 5 years) and vaginal parity. Matching resulted in groups of similar age (controls 53 ? 13.3 years compared with cases 55 2 12.4 years, P > .05) and parity (2.3 ? 1.5 compared with 2.9 2 1.6, P > .05). Paired t test was used to assess the significance of the difference between these matched groups.
Results Support of the pelvic organs, group-specific demographics, and previous prolapse surgery history are shown in Table 2. The area of the urogenital hiatus in the levator ani among women not previously operated on is shown in Figure 2. Of the patients who had significant prolapse (grades 2 and 3), only 26% fell within the normal range (mean ? 2 SD; ~8.4 cm’), and only 6% were at or below the mean of 5.4 cm’. The case
Figure 1. Dimensions measured of the urogenital ani muscles. Al’ = anterior-posterior diameter; diameter; PB = perineal body.
DeLancey
and
Hurd
hiatus Trans.
in the levator = transverse
Urogenital Hiatus and Prolapse
365
Table
2. Support
Status
and
Previous
Surgical
No prior Grade
n
prolapse
Age
0
197
39
1
27
46
2 3
18 16
52 61
Data are presented
as mean
k standard
Treatment
for
Prolapse
surgery
Prior
(y)
n
Parity
t 11.8 + 8.8 -c 13.4 -c 16.5
1.3 2.2 2.3 3.3
2 1.2 -c 1.5 ? 1.75 2 1.7
prolapse
surgery Parity
Age (y)
6
56 t 63 5 56 ir 64 2
10 12 14
12.8 16.1
3.2
10.8 16.1
2.4 3.5
2.1
k 2 t 2
1.5 1.4 1.2 1.5
deviation.
control analysis carried out to assess the possibility that age or parity was responsible for difference in hiatus size showed a persistent difference between cases with unrepaired prolapse (mean hiatus area 9.5 cm’ 2 3.8) compared with normal support controls (mean area 5.7 cm’ i 1.4, P < .05). The relationship between prior surgery and prolapse, and size of the hiatus is shown in Figure 3. Among patients with recurrent prolapse, those who had two or more previously failed attempts to correct prolapse had a larger hiatus than those in whom surgery had been successful (P < .05) or those who had only one previous failed repair (P < .05). Even in patients whose prolapse had been cured surgically, the hiatus did not return to normal size (P < .05). The increase in hiatus size was associated more strongly with the changes in the anterior-posterior diameter (? = .9, P < .05) whether measured to the perineal body or the anus (Figure 4). There also was .some increase in transverse diameter as the area enlarged (Y* = .6, P < .05) (Figure 5). There was no correlation between the thickness of the perineal body and the size of the urogenital hiatus (r2 = .0002, P > .05) (Figure 6).
Discussion This study revealed that the size of the urogenital hiatus is larger in women with prolapse than it is in women with normal support. This difference persisted even in those women who have been cured of their prolapse. The enlarged hiatus, therefore, is not entirely a reflection of the presence of the prolapse. Furthermore, the hiatus is progressively larger in women with increasingly severe prolapse and is largest in those individuals in whom several surgical repairs have failed to correct the prolapse. The degree of prolapse present, rather than a women’s presenting complaint, was used in this report because the study’s intent was to correlate changes in size of the hiatus with differing degrees of loss of support. In each woman, the type of prolapse was noted (eg, enterocele and cystocele). Because of the great diversity of findings, this information was not used for analysis, because each group was too small to yield reliable information.
16
16
r
T
L
2
t-
0
0 1
ti
0
1
II=197
n=27
283 n=18
n=16
and
Hurd
Urogenital
-
Failed 1 Oper. n=16
Cured n=16
i
Failed > 1 Oper n=8
Surgical History
Figure 2. Hiatus size in women not operated on previously for prolapse, by grade with standard deviation shown. All clinical prolapses (grades 2 and 3) are summarized at right. *P i .05 compared with normal.
DeLancey
Unrepaired n=34
II=34
Grade
366
Normal n=198
Hiatus
and Prolapse
Figure 3. Influence of prior surgery and its success on hiatus area, with standard deviation shown. Oper. = operation. *P < .05 compared with normal. **Compared with unrepaired, cured, and after one failed repair.
Obstetrics
& Gynecology
8
y = 8.9E-3x
+ 2.3,
r2 = 1.9E-3
7
Perineal Body Thickness
--Y.-+-,
2.5
0
5
7.5
hlarus
__I-
12.5
15
Size
(cm.
sq.)
17.5
20
22.5
.Q
2.5
5
7.5
Hiatus
10
Area
12.5
15
17.5
20
2
(cm. sq.)
4.
Although the firtdilkgs ut this study established an aasociatlon between Moe of the hiatus and pelvic organ prulapse, they did not establish causation, determine the structure that is responsible fur hiatus closure, or uldicate why closure should influence pelvic organ support. Kelly’ observed that it is the levatur ani and Ilut the pttli~~al body ur bulbucdvernous muscle that &ses the vagmdl outlet, and our anatomic observations were corkstent with this observation. The direction of
gr--,-.---’
. ’ N-----t Aws
’
/
/
/. / 6-
rtarlsv~f5~ Diameter (cm.1
,
/
/
/
/
/
,
F-b / / t
1 /-
t
1 /
0 1 .0
I -.--.. 2.5 5
iv 7.5
Hiatus ~~gurr 5. hxieidtlul~ hi&us and the
YI,
NO.
brmern
vwa.tw 10 12.5
Size
15
17.5
20
22.5
(cm. sq.)
the t1d11svelse didmeter of the urogeahiatus. [
VUL
!
v_-
10
~u~~elat~u~l between the sqgttdl dlanneters of the urogenital h~us mrusured frum the pubic borle to either the perineal body (I’S, +J)L,I LL&SJ or the uus (oyei~ burrs) and the ared of the hiatus. (All pktts x+zh sm&u medsureinents are represented by d s~rrgle dot.)
kq+z~e
6
Dora of the
5, MAKLH
1998
Figure 6. Correlation between the thickness of the perineal body the area of the hiatus. (All patients with the same measurements represented by a single mark.)
and are
the muscle fibers within the levator ani muscle are such that muscular relaxation would result in an increase in the distance between the pubic bone and the perineal body and anus. Furthermore, the constant contraction of the levator ani muscle’ is well suited to maintaining continuous closure of the pelvic floor and is similar to the continuous tone in the external anal sphincter. Our data indicating increased anterior-posterior diameters but no change in perineal body thickness further supported Kelly’s concept. The hypothesis that damage to the levator ani muscle plays a role in causing pelvic organ prolapse has been proposed for many years. Direct dissections of these muscles in cadavers with prolapse’ and radiographic studies2 have demonstrated this role. More recently, electromyography has identified partial denervation of the levator ani muscles as being associated with prolapse3 and as possibly explaining the loss of function noted in the earlier studies. Halban and Tandler’s direct anatomic studies’ of the pelvic floor in women with prolapse concluded that injury to the pelvic floor with opening of the so-called pelvic valve was instrumental to developing prolapse. Berglas and Rubin observed that the levator plate loses its horizontal position in patients in association with enlargement of the levator hiatus, and Nichols et al7 pointed out the clinical implications of this change. Because changes in the closure of the urogenital hiatus would be expected to be associated with changes in the inclination of the levator plate, these two changes probably are reflections of the same phenomenon. The finding that an enlarged hiatus is associated with
DeLancey alld Hurd
Ur.ogmitd
Hlutus
ad
Prolapse
367
genital prolapse suggested the following hypothetical mechanism by which damage to the levator ani muscle and enlargement of the urogenital hiatus might influence pelvic support. In the normal individual, the 5-cm* area of the hiatus is filled with the vaginal walls, endopelvic fascia, and urethra. In effect, these organs fill the opening in the muscles, forming a seal in its opening. As the size of the opening within the muscle enlarges, an open space in the pelvic floor at the vaginal opening develops. Because the vaginal canal does not contain air, something must fill this space. This may be a cystocele, rectocele, or the uterine cervix. Once the pelvic outlet has opened and the vaginal wall or cervix lies unsupported at the boundary between abdominal and atmospheric pressure, the endopelvic fascia is called on to attempt to keep the unsupported pelvic organs in place, unsupported by the levator ani muscles. The constant load of abdominal pressure on this connective tissue over time could be expected to cause failure of the connective tissue in certain predisposed individuals, with the development of significant prolapse.8 It should be noted that it is not necessary for the levator ani muscles to be damaged for prolapse to occur, because a few women in our study have overt prolapse despite a normal hiatus. It is possible for the cervix or vaginal wall to prolapse downward because of defective connective tissue supports without damage to the muscles. In this situation, the prolapse simply dilates an otherwise normal pelvic floor by driving a wedge through the urogenital hiatus. This explains the overlap in hiatus size between normal women and some women with prolapse. Nevertheless, the vast majority of women with prolapse have a hiatus that falls outside the normal range, indicating that prolapse in women with a normal hiatus is the exception rather than the rule. Important questions remain to be answered in determining the etiology of pelvic organ prolapse. The factor responsible for enlargement of the urogenital hiatus remains to be determined: Is it neurologic injury to the levator ani muscles, or is it direct muscle trauma as might occur at the time of vaginal birth? What is the
368
DeLancey
and
Hurd
Urogenital
Hiatus
and Prolapse
relationship between abnormally weak connective tissue and these changes in hiatal size? How do pelvic muscle exercise and surgical intervention influence the size of the urogenital hiatus? Now that a simple technique is available to assess urogenital hiatus size, it should be feasible to begin to gather data concerning these issues.
References 1. Halban J, Tandler J. Anatomie und atiologie der genital prolapse beim weibe. Vienna and Leipzig: Wilhelm Braunmuller, 1907. 2. Berglas B, Rubin IC. Study of the supportive structures of the uterus by levator myography. Surg Gynecol Obstet 1953;97:677-92. 3. Smith ARB, Hosker GL, Warrell DW. The role of partial denervation of the pelvic floor in the aetiology of genitourinary prolapse and stress incontinence of urine: A neurophysiological study. Br J Obstet Gynaecol 1989;96:24-8. 4. Baden WF, Walker T. Fundamentals, symptoms, and classification. In: Baden WF, Walker T, eds. Surgical repair of vaginal defects. Philadelphia: JB Lippincott, 1992. 5. Kelly HA. Gynecology. New York: Appleton and Company, 1928: 265-7. 6. Parks AG, Porter NH, Melzak J. Experimental study of the reflex mechanism controlling muscles of the pelvic floor. Dis Colon Rectum 1962;5:407-14. 7. Nichols DH, Milley I’S, Randall CL. Significance of restoration of normal vaginal depth and axis. Obstet Gynecol 1970;36:251-6. 8. Paramore RH. The supports-in-chief of the female pelvic viscera. J Obstet Gynecol Br Emp 1907;13:391-409.
Reprints are not available. Address correspondence to: John 0. L. DeLancey, MD Department of Obstetrics and Gynecology University of Michigan Medical School L4100 Women’s Hospital 1500 East Medical Center Drive Ann Arbor, MI 48109-0216 E-mail: [email protected]
Received Received Accepted
May 22, 1997. in revised form September September 10,1997.
Copyright 0 Gynecologists.
1998 by The American Published by Elsevier
9, 1997.
College of Obstetricians Science Inc.
Obstetrics
b Gynecology
and