This prospective study was approved by the local ethics committee (Study-ID: BASEC 2018-01107), and all patients and volunteers gave written informed consent.

Twenty-two consecutive volunteers and 25 consecutive patients from the urogynecology unit of our hospital were enrolled.

Patients were included, if they had symptoms of POP as assessed with a standardized questionnaire [13] and a stage 2 or more prolapse in any compartment as assessed through urogynecologic examination using the clinical “Pelvic Organ Prolapse Quantification (POP-Q)” system [14]. The POP-Q system ranges from 0 (no prolapse) to 4 (maximum descent) scored according to the extent of organ prolapse relative to the hymen as the anatomic reference point [15].

The standardized and validated pelvic floor questionnaire that was used [13] integrates four domains of pelvic floor dysfunction (bladder function, bowel function, sexual function, and pelvic organ prolapse), grades their severity, and assesses bothersomeness and condition-specific quality of life. For each of the four domains, a total value will be returned and calculated into a total “pelvic floor dysfunction score” which can reach a maximum of 40 points. A higher score correlates with increased severity of pelvic floor dysfunction.

Volunteers were included if symptoms of pelvic organ prolapse were absent (assessed through a structured interview prior to inclusion and with the standardized questionnaire) and if they had never given birth (nulliparity).

Additional inclusion criteria comprised obtaining written informed consent and the complete MRI exam according to study protocol. Exclusion criteria were inability to follow the instructions during the MR image acquisition, failure to return the questionnaire or undergo physical examination (for patients only), history of prior pelvic floor surgery, insufficient straining maneuver during the dynamic phase of the MRI, and general contraindications to MRI (e.g., presence of non-MR-compatible metallic implants, devices or metallic foreign bodies). A part of this cohort was previously used for another study with a different research objective that focused on comparison of different techniques for acquisition of dynamic MRI sequences [16].

MRI examinations were performed on a 3.0 T clinical MRI scanner (Skyra, Siemens Healthineers, Erlangen, Germany) with a 60-channel array coil. Subjects were examined in supine body position and emptied their bladder 15 min prior to the exam. Prior to image acquisition, the participants were instructed by the MR technologist on how to perform the straining maneuver for the dynamic phases of the examination and were allowed to practice the maneuver before the image acquisition. In this study, we analyzed the dynamic midsagittal single-slice sequences obtained at rest and straining (true fast imaging with steady state free precession [TRUFI], TR/TE, 460/1.5 ms; matrix, 320 × 320; FOV, 240 × 240 mm; slice thickness, 10 mm). Images were acquired during three consecutive straining maneuvers (total acquisition time 1 min 10 s, 72 consecutive images per straining maneuver).

One radiologist (S.G., 7 years of experience in abdominal imaging) annotated the organ points of the bladder (anterior compartment), cervix (middle compartment), and anorectal junction (posterior compartment) at rest and straining in volunteers and patients. The straining phase depicting the maximum pelvic organ descent was used for measurements. Annotations were done on an in-house developed tool called “3D PICS” [11]. Briefly, this tool is based on a 3D coordinate system using pre-defined bony landmarks (the inferior margin of the symphysis, the sacrococcygeal joint, and the ischial spines) and uses the PICS plane as a reference line. The PICS plane is drawn as a line with 34 degrees (for images at rest) or 29 degrees (for images at straining) clockwise rotation from the “sacrococcygeal inferior—pubic point line” (SCIPP line) (Fig. 1). The SCIPP line is a line connecting the inferior margin of the symphysis to the anterior sacrococcygeal joint. The need for adjustments in rotational angles (i.e., 34 and 29 degrees) between images at rest and straining is informed by earlier research demonstrating variations in pelvic tilt between these states [5]. To account for these differences and ensure accurate measurements, the line is modified accordingly depending on whether measurements are made on resting or straining images. The coordinates of any given point annotated on MR images can be calculated using 3D PICS. When using single-slice midsagittal TRUFI images, two coordinates are calculated (position along the y-axis in the craniocaudal direction, and the position along the x-axis in the antero-posterior direction) [11]. For better comparison with the PCL, we analyzed the position in the y-axis (craniocaudal direction) only. The PCL represents a straight line connecting the inferior border of the pubic symphysis to the last coccygeal joint and is the most widely used reference line for the grading of POP on MRI [3].

38-year-old (gravida 2, para 2) patient with multi-compartment organ prolapse presenting with symptoms of vaginal bulge and stress urinary incontinence. Top row depicts dynamic midsagittal MR images showing the position of the PICS line at rest (A) and straining (B) adjusted for differences in pelvic inclination. The interrupted line represents the sacrococcygeal-inferior pubic point (SCIPP) line which connects the sacrococcygeal joint to the inferior border of the pubic symphysis. The PICS line (solid line in A–D) is then drawn from the inferior border of the pubic symphysis with a 34° and 29° clockwise rotation from the SCIPP line at rest and straining, respectively. Bottom row (C, D) shows organ point measurements with reference to the PICS line in this patient at rest (C) and straining (D). At rest, the bladder (Bl), cervix (Cx), and anorectal junction (ArJ) are located at 1.5 cm, 4.2 cm, and 0.9 cm above the PICS line. At straining, the bladder descents to 4.0 cm below, the cervix to 1.9 cm below, and the anorectal junction to 3.1 cm below the PICS line. Measurements are made perpendicularly to the PICS line. Clinically, the patient was graded as POP-Q III° for the anterior, POP-Q II° for the middle, and POP-Q I° for the posterior compartment

The 3D PICS tool is based on organ point locations measured perpendicularly to the PICS line. For the manual measurements of the organ point positions relative to the PCL, the measurements were also done perpendicularly to the PCL [3]. For the PCL and PICS measurements alike, coordinates of organ points located above the reference line are labeled with a negative sign and those below the reference line with a positive sign (Fig. 1).

In addition, images were reviewed in the Picture communication and archiving system and the position of the three organ points was measured at rest and straining in reference to the PCL. To define prolapse based on PCL measurements, we used established thresholds that were published in the joint consensus recommendations of the ESUR and ESGAR Pelvic Floor Working Group [3].

Categorical variables were described as frequencies and percentages. Continuous variables were described as means and standard deviations or medians and interquartile range where appropriate.

Demographics of the study population and organ point measurements at rest and straining in patients and volunteers were analyzed using descriptive statistics. Differences in age and body mass index between patients and volunteers were tested with the Mann–Whitney U test.

Pelvic organ point measurements were tested for normal distribution using the Kolmogorov–Smirnov test.

Differences in organ positions at rest and straining between patients and volunteers were tested with a student’s t-test or Mann–Whitney U test. In patients, differences in the proportion of different POP-Q stages for the anterior, middle, and posterior compartment were tested with a McNemar-Bowker test.

Receiver-operating characteristic (ROC) curves with calculation of the area-under-the-curve (AUC) were plotted to identify compartment-wise cutoff values for pelvic organ point measurements at straining for diagnosis of pathologic pelvic floor descent. The 95% confidence intervals (CI) were computed using a bootstrapping approach. Following thresholds were applied to assess the discriminative power using the AUC: < 0.5 = not useful, 0.5–0.6 = poor, 0.61–0.7 = acceptable, 0.71–0.8 = good, 0.81–0.9 = very good, and 0.91–1.0 = excellent [17]. Cutoff values for organ point positions from the ROC analysis were set to yield high sensitivity and specificity for the differentiation between patients (with POP-Q stage ≥ 2 in the respective compartment) and healthy volunteers. In addition, the sensitivity and specificity of PCL measurements for differentiation between patients and volunteers were calculated using established threshold values for the PCL at straining [3].

The correlation between PICS measurements and PCL measurements and compartment-wise POP-Q stage was tested with Spearman’s rank correlation. The Spearman’s coefficient (δ) ranges from − 1 to + 1, where 0 indicates that there is no linear association and 1 indicates perfect linear association [18]. The following thresholds were applied to assess the strength of relationship: < 0.2 = no or negligible relationship, 0.20–0.29 = weak relationship, 0.30–0.39 = moderate relationship, 0.40–0.69 = strong relationship, and ≥ 0.70 = very strong relationship [19]. Differences in sensitivity and specificity between PICS measurements and PCL measurements were compared with a McNemar test.

A 2-tailed p value of < 0.05 was used to determine the statistical significance. Statistical analyses were performed with SPSS (version 29, IBM Corporation, Armonk, NY, USA).