Article Text
PDF
Abstract
Backgrounds Pelvic organ prolapse commonly affects the upper two-thirds of the vagina. However, evaluating this region in its normal position presents challenges. This study aimed to assess the anatomical axis of this vaginal segment using pelvic magnetic resonance imaging measurements.
Methods A retrospective study of 614 hospitalized women from two hospitals, who were rigorously screened to exclude those with Pelvic Organ Prolapse or known anatomical variations (median age: 43 years, range: 17–76 years). Two reference lines were used: the pubococcygeal line (PCL) and a line from the inferior pubic symphysis to the third sacral vertebra (PS3L). Distances between the distal, middle, and apical points of the upper vagina and the reference lines, as well as the angles between the upper vagina and the reference lines, were measured. Comparisons were made among different age groups.
Results The median distances from the distal, middle, and apical vaginal points to the PCL were 0.4 cm (interquartile range [IQR]: 0.0–0.7 cm), 2.1 cm (IQR: 1.7–2.5 cm) and 3.1 cm (IQR: 2.5–3.7 cm), respectively. The median PCL-vaginal angle was 29.0° (IQR: 23.0–34.0°). The median distances from the distal, middle, and apical vaginal points to the PS3L were -0.5 cm (IQR: -0.9–0.0 cm), 0.0 cm (IQR: -0.4–0.6 cm), and -0.2 cm (IQR: -0.9– 0.0 cm), respectively. The median PS3L-vaginal angle was 0.0° (IQR: -4.0–7.0°). Women aged 50 years or older had slightly lower vaginal points and slightly larger angles than younger groups in relation to both reference lines (p < .001). The vaginal axis in younger groups appeared parallel to the PS3L.
Conclusions The axis of the upper two-thirds vagina was proximate to a line from the inferior pubic symphysis to the third sacral vertebra, particularly in younger women. It will likely become a simplified method for roughly assessing the vaginal axis in its situ at first glance.
Highlights
Pelvic organ prolapse commonly affects the upper two-thirds of the vagina; evaluating this region in situ is challenging.
This retrospective study used pelvic MRI measurements to assess the normal axis of the upper two-thirds of the vagina.
A novel line passing through the inferior portion of the pubic symphysis to the third sacral vertebrae as PS3L was introduced.
Three Marker points of vagina relatively concentrated along PS3L, vaginal plane nearly parallel to PS3L.
The findings may allow the assessment of the upper two-thirds vaginal axis simple and easy.
- Levator ani muscle
- Pelvic organ prolapse
- Magnetic resonance imaging
- Pubococcygeal line
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Statistics from Altmetric.com
1 Introduction
The vagina, as one of the three outlets of the female pelvic floor, serves as the largest passageway and facilitates childbirth. Evaluating the normal position of the vaginal axis is of paramount importance. For instance, pelvic organ prolapse (POP) is a common and distressing condition characterized by the descent of the vaginal walls.1 The lower third of the vagina, known for its dense attachments, exhibits a lower susceptibility to POP, whereas POP commonly affects the upper two-thirds of the vagina.2 A clear understanding of the normal anatomical appearances of the upper two-thirds of the vagina in its natural position is a prerequisite for accurately assessing pelvic organ prolapse.
Currently, quantifying the orientation and position of the vagina requires establishing a two-dimensional or three-dimensional coordinate system.3,4 This approach is complex and challenging for clinicians and radiologists, making it difficult to master and limiting its widespread clinical applicability. In clinical practice, rapid assessing vaginal orientation is valuable. However, despite various proposed reference lines, none of them represent the plane of the upper vagina.5
In magnetic resonance imaging (MRI), the pubococcygeal line (PCL) is the most frequently referenced for evaluating POP. It represents the level of the levator ani muscle (LAM).5,6 The LAM is vital in supporting the pelvic organs, with the upper two-thirds vagina resting upon it.7 Therefore, it is likely that the axis of the upper vagina is in close proximity to the PCL. However, there are additional supporting structures in the upper segment of the vagina. Anteriorly, the upper vagina merges with the fascia of the LAM and terminates at the inferior border of the pubic symphysis through the pubocervical fascia. Posteriorly, it is supported by the uterosacral ligament, which attaches to the S2-S4 segments of the sacrum.8 Based on initial experimentation, we speculate that the axis of the upper vagina if determined by bony landmarks, is likely to extend from the inferior border of the pubic symphysis to the middle portion of the third sacral vertebra in the sagittal plane (Fig. 1A). The primary objective of this study is to confirm whether this line well estimates the upper two-thirds vaginal axis with a large sample.
Sagittal plane images of the parameters.
A, sagittal MRI image of the pelvic floor was obtained in a 29-year-old primiparous woman diagnosed with pelvic inflammatory disease. It was performed with a fat-saturated T2-weighted fast recovery spin-echo; no contrast material was used.
B, bladder base; C, the most distal edge of the cervix; D, the apex of the tented up posterior fornix; PS, pubic symphysis; P, the most inferior aspect of the PS. PCL: a pubococcygeal line is drawn from point P to the last coccygeal joint. PS3L, a line passing through point P to the midpoint of the third sacral vertebrae. A perpendicular line was drawn from the bladder neck (Point B) to the PS3L, and this line was extended into the vaginal area, and the midpoint of the vaginal portion (green portion in Fig. A) of this line was designated as Point b.
A line through points b and D represents the upper vagina's axis (Fig. C). The β, θ (red color) represent the PS3L-vaginal angle and PCL-vaginal angle, respectively (Fig. C).
2 Materials and methods
This retrospective cross-sectional study received approval from the ethical committee (Approval No. K20190046). Informed consent requirements were waived. Demographic information was extracted from the medical records.
Inclusion Criteria: A comprehensive electronic search of the radiology information system was conducted in two hospitals (The Fourth Affiliated Hospital of Zhejiang University School of Medicine and the Women's Hospital, School of Medicine, Zhejiang University) from July 2018 to June 2019, encompassing all consecutive female inpatients who underwent pelvic MRI.
Exclusion Criteria: To exclude those with POP or other potential anatomical variations, we implemented strict exclusion criteria by carefully reviewing each patient's history, ultrasound or MRI image reports, and gynecological examination records from medical records. Additionally, we reviewed the surgical records available. As teaching hospitals, all patients with POP have undergone POP-Q assessment, and these records have been included in their medical records. However, for patients without POP, POP-Q records were not available. Patients who met one of the following criteria were excluded: (1) cases with POP based on pelvic examination records; (2) a history of prolapse or urinary or fecal incontinence; (3) previous pelvic floor surgery for prolapse or urinary incontinence;(4) a previous history of total or subtotal hysterectomy or cervical excision; (5) pregnancy with a gestational age ≥ eight weeks or within 12 months after delivery; (6) malformation of the female genital tract, such as Mayer-Rokitansky-Küster-Hauser syndrome, although uterine mediastinum was not excluded; (7) a maximum uterine diameter ≥ 7 cm, such as multiple uterine fibroids or adenomyosis; (8) any mass or cysts in the pelvis with a diameter ≥ 6 cm, such as fibroids, ovarian cysts, or hydrosalpinx; (9) any mass or cysts in the cervix or the cervical canal with a diameter ≥ 2 cm; (10) any mass or cysts on the vaginal wall, in the vaginal lumen, or the area around the vagina with a diameter ≥ 2 cm; (11) a maximum diameter of the rectum ≥ 3 cm or maximum bladder diameter ≥ 6.5 cm on any sagittal images; (12) poor visibility of the anatomic structures of the landmarks; and (13) for patients with surgical records available, those had dense, extensive adhesions between the uterus and abdominal wall suggestive of an obvious change in the position of the uterus.
2.1 MR image analysis
Pelvic MRI scans were conducted in both of our hospitals using a 1.5-T magnet with patients in the supine position. The measurements from the MRI scans were obtained using WebViewer (Greenlander Information Technology, Version 1.0.0.53449, Hangzhou, China) on the sagittal plane, either at the midline or close to the midline. A fat-saturated T2-weighted fast recovery fast spin-echo sequence was utilized, with a repetition time (TR) of 4100–4200 ms and an echo time (TE) of 102–109 ms. The measurements were performed independently by a gynecologist-obstetrician and a radiologist, both blinded to the study hypothesis.
The sagittal diagram of the pelvis in Fig. 1 illustrates several key anatomical landmarks. Point P corresponds to the most inferior aspect of the pubic symphysis, while point C indicates the most distal edge of the cervix. Point D represents the apex of the tented-up posterior fornix. Two distinct reference lines were employed (Fig. 1A and B). The first line, PS3L, was drawn from point P to the midpoint of the third sacral vertebra. The PCL was the second line from point P to the last coccygeal joint.6
A perpendicular line was drawn from the bladder neck (Point B) to the PS3L, and this line was extended into the vaginal area, and the midpoint of the vaginal portion of this line was designated as Point b. Points b, C, and D were expected to represent the distal, middle, and apical regions of the upper vagina, respectively. The perpendicular distances from Points b, C, and D to both reference lines were measured (Fig. 1B).
Furthermore, the angles between the axis of the upper vagina (a line connecting point b to point D) and both the PS3L and PCL, as well as the angle between the PCL and PS3L, were determined (Fig. 1C). All measurements were recorded in centimeters or degrees. If an anatomical landmark was positioned above a reference line or an angle rotated counterclockwise, it was assigned a positive value. Conversely, if an anatomical landmark was situated below a reference line or an angle rotated in the clockwise direction, it was assigned a negative value.
2.2 Statistical analysis
The data analysis was conducted using IBM SPSS Statistics version 23. The normality of distribution for continuous variables was assessed using the Kolmogorov-Smirnov test. Descriptive variables like baseline data and MRI outcomes are presented as percentages (n (%)), median (interquartile range (IQR)), and mean (standard deviation (SD)). Spearman's correlation coefficient was used to demonstrate the potential effect of influencing factors (age, BMI, parity, previous abdominal surgery, and pregnancy) on the position of vaginal points referring to PS3L. To better observe the sample, we conducted a rough age stratification based on menopausal age and parity. And found that cases below the age of 30 were predominantly those who had not completed their reproductive journey, individuals between 30 and 49 years were considered within the reproductive age group, and 50 years was the average age of menopause. So we stratified the patients into three categories:≤ 30 years, between 31 and 49 years, and≥ 50 years.Between-group comparisons of continuous data were performed using a two-independent t-test or a Mann-Whitney U test, while differences in categorical data were evaluated using a chi-square test. A significance level of P < .05 was considered statistically significant.
3 Results
3.1 Patient characteristics
A total of 3,545 consecutive patients who underwent pelvic MRIs were included in the initial review. Among these patients, 2,931 individuals met the exclusion criteria, leaving 614 cases for the final analysis (Fig. 2). Of the 614 cases, 333 patients underwent MRI for benign and malignant uterine lesions, 179 for adnexa diseases, 80 for cervix lesions, and 22 for other factors such as vulvar diseases and abdominal wall conditions. The median age of the study sample was 43 years (range: 17–76 years), and the median body mass index was 22.4 kg/m2 (range: 15.6–38.8 kg/m2). Among the patients, 390 (63.5 %) had laparoscopic or transabdominal surgery records.
Exclusion criteria and number of participants excluded.
3.2 Assessment of the vaginal axis using PCL and PS3L
The median distances from the distal, middle, and apical points of the upper vagina to the pubococcygeal line (PCL) were 0.4 cm [IQR, 0.0–0.7 cm], 2.1 cm [IQR: 1.7–2.5 cm] and 3.1 cm [IQR: 2.5–3.7 cm], respectively (Table 1). The PCL and the vaginal axis formed an acute angle with a median of 29.0° [IQR: 23.0–34.0°].
Demographic characteristics and MRI measurements.
The median distances from the distal, middle, and apical points of the upper vagina to the line connecting point P to the midpoint of the third sacral vertebra (PS3L) were -0.5 cm [IQR: -0.9–0.0 cm], 0.0 cm [IQR: -0.4–0.6 cm], and -0.2 cm [IQR: -0.9–0.0 cm], respectively (Table 1). The PS3L was nearly parallel to the vaginal axis, and the median angle between the PS3L and the vaginal axis (PS3L-vaginal angle) was 0.0° [IQR: -4.0–7.0°] (Table 1).
3.3 Vaginal axis in different age subgroups
Correlation analyses for the relative positions of the three studied points concerning PS3L in relation to age, BMI, parity, vaginal delivery, cesarean section, and previous abdominal surgery showed that only the relative position of the b to PS3L exhibited a weak negative correlation with age (r = -0.49, p < .001), parity (r = -0.36, p < .001), and vaginal delivery (r = -0.40, p < .001) (Table 2). Age exhibits a positive correlation with parity and vaginal delivery (p < .001).
Spearman's correlation coefficients for age, BMI, parity, VD, CS, PAS and the position of vaginal points relative to PS3L.
There were 120 patients aged ≤ 30, 282 between 31 and 49, and 212 patients aged ≥ 50. Body mass index (BMI) values were generally low in all three subgroups, with limited differences observed among the groups. In the oldest group, 92.9 % (197/212) of patients had experienced vaginal delivery, with 49.1 % (104/212) multiparous. Conversely, in the youngest group, 79.2 % (95/120) were nulliparous (Table 1).
The vaginal marker points in the oldest group were slightly lower than those in the younger groups, whether referring to the PCL or the PS3L (p < .001) (Table 1 and Fig. 3). Moreover, both the PCL-vaginal angle and the PS3L-vaginal angle were slightly larger in the oldest group compared to the two younger groups (p < .001) (Table 1). The vaginal axis in the youngest groups more closely conformed to the PS3L, appearing parallel to it with the marker points nearly aligned along the line, resulting in median PS3L-vaginal angles of zero degrees (Table 1 and Fig. 4).
Vaginal axes of different age groups.
Points D, C, and b represent the apical, middle, and distal points of the upper vagina. Each line represents one woman. Line 0 refers to the reference line (PS3L or PCL). The purple represents measurements referring to PS3L (Fig. 4A–C), while the red refers to PCL (Fig. 4D–F). Above a reference line, it has a positive negative value, while below it a negative value.
Violin Plots: Distances from Vaginal Marker Points to PCL and PS3L
Points D, C, and b represent the apical, middle, and distal points of the upper two-thirds of the vagina. The 0 line refers to the reference line (PS3L or PCL). Solid lines in the violin plots represent the median value, while dotted lines in the violin plots represent the 25th and 75th percentiles of value, respectively. Above a reference line, it has a positive negative value, while below it a negative value.
4 Discussion
4.1 Main findings
This hospital-based cross-sectional study focused on the sagittal plane orientation of the upper two-thirds vagina, excluding individuals with prolapse or known anatomical variations. The findings revealed that the axis of the upper vagina was positioned approximately 2–3 cm above the levator ani plane represented by the pubococcygeal line (PCL), forming a 29-degree angle with the PCL. Furthermore, the vaginal axis is closely aligned with a line extending from the inferior border of the pubic symphysis to the third sacral vertebra (PS3L). The median distances from the vaginal marker points to the PS3L ranged from −0.5 cm to 0.0 cm, and the median angle between this line and the vaginal axis was 0.0°. These observations were consistent across all age subgroups, with the younger groups showing greater conformity to the PS3L.
4.2 Interpretations
In this study, the MRI was not done under maximal Valsalva. Although prolapses are assessed during maximal Valsalva, this study is to find a normal vaginal axis, and the pelvic structure is more likely to be in situ under rest. When assessing the position of the vagina, it is crucial to understand its supporting structures. The vagina was categorized into three levels according to DeLancey's classification.2 The cardinal and uterosacral ligaments commonly support the upper third of the vagina; the middle third is laterally attached to the arcus tendineus fasciae of the pelvis; the lower-third merges with the urethra and attaches to the pubocervical fascia.2 However, recent evidence has shown that the cardinal ligament functions as a mesentery structure comprising vessels and nerves.8–10 Thus, the uterosacral ligament and pubocervical fascia are the primary ligament-like structures supporting the upper two-thirds vagina. In other words, it fuses at the inferior border of the pubic symphysis anteriorly and ends at the S2-S4 sacral segments posteriorly, with the axis closely approximating the PS3L. However, the measurements still showed variation in these parameters, possibly because the uterosacral ligament is attached anteriorly mostly but not exclusively on the posterior aspect of the cervicovaginal junction, posteriorly broadly to the first three sacral vertebrae and variably to the fourth sacral vertebra.11,12 In addition, the definitive role of paravaginal support in the middle third of the vagina and its contribution to the development of prolapse are still unknown.13,14
The study population comprised individuals of varying ages, parity, BMI, and delivery types. Additionally, many participants had undergone abdominal surgical interventions, and some patients had missing surgical records. However, it is essential to note that correlation analyses showed that only the relative position of the bladder neck to PS3L exhibited a weak negative correlation with age, parity, and vaginal delivery. It has been reported that aging, multiparty, and especially previous vaginal delivery are high-risk factors for pelvic organ prolapse.15 Generally, the three vaginal marker points are well followed by the PS3L in all age subgroups, especially in young nulliparous women. In the multiparous that was over 50 years old but had no POP history and were not diagnosed as POP, the vaginal axis was slightly lower. A moderate degree of prolapse in continent women based on MRI has been reported due to its ability to measure actual pelvic organ descent.16,17
4.3 Limitations
Our study has several limitations. Ideally, for studying the natural axis of the normal vagina, prospective recruitment of healthy, young, nulliparous women for MRI would be the ideal approach. However, due to limitations in research funding, our study utilized a retrospective study involving MRI cases from hospitals. To address this limitation, we implemented strict exclusion criteria. Through these efforts, we aimed to exclude potential cases with abnormal pelvic organ positions to the best of our ability, ensuring a relatively “normal” sample for analysis. However, based on the current literature review, there is no evidence to suggest that those exclude potential factors would influence the vaginal axis. The upper limits of the measurements are primarily based on our observations during an extensive review of imaging cases. The retrospective design of the study introduces inherent limitations. Secondly, the MRI scans were conducted with patients in the supine position, which may not fully reflect the positions of pelvic organs in the standing position. This positioning limitation should be considered when interpreting the results. Lastly, the demographics of the study population, including the relatively low BMI and the specific ethnic group (Mongolian), may restrict the generalizability of the findings to other populations. Further studies involving a more diverse population and a normal sample are warranted to validate our findings.
5 Implications
This hospital-based cross-sectional study elucidates the typical anatomical orientation of the upper two-thirds of the vagina on magnetic resonance imaging in women without pelvic organ prolapse. By introducing a novel reference line from the inferior pubic symphysis to the third sacral vertebra, the findings provide a simplified method for radiographic assessment of the upper vaginal axis, with potential implications for evaluating pelvic organ prolapse and guiding surgical reconstruction. Notably, the study demonstrates close concordance between the upper vaginal axis and this reference line, particularly among younger nulliparous women. These normative data substantiate this radiographic measure as a useful indicator of normal vaginal anatomy and orientation. Although limited by the retrospective design and population characteristics, these findings may facilitate pelvic floor disorder diagnosis, preoperative planning, and postoperative evaluation in a clinical setting. Further prospective study in a larger, more diverse cohort is warranted to validate the reliability of this reference line for research and clinical applications.
6 Conclusions
Our study found that the upper two-thirds vagina mostly conforms to a reference line passing through the most inferior aspect of the pubic symphysis to the midpoint of the third sacral vertebra, especially in the younger nulliparous women. It may allow the assessment of the upper vaginal axis simple and easy.
Author contributions
OY is responsible for Conceptualization, methodology, formal analysis, investigation, writing-original draft. LF,WR, XW, ZW, and YW are responsible for formal analysis, and investigation. ZX is responsible for conceptualization methodology, writing-Review& editing, supervision, and project administration.
Funding
Financial support for this project was provided by the Basic Public Welfare Research Project of Zhejiang Province (Project No: LGF22H040018), a nonprofit organization.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
Thanks very much to Dr. Yongqing Zhang, my colleague, for his warm help in statistics and revision.
We thank Mr. Yuyun Zhou, an excellent architect, for helping us make the schematic look good.