Patient and volunteer characteristics

Two symptomatic patients were excluded (unable to follow instructions during the MRI exam and incomplete exam n = 1, failure to return the questionnaire and undergo urogynecologic examination for POP-Q grading n = 1). The final population comprised 23 consecutive premenopausal patients with symptomatic POP and 22 asymptomatic premenopausal nulliparous volunteers. Baseline characteristics of the patients and volunteers are depicted in Table 1.

Table 1 Baseline characteristics of patients and healthy volunteers

Differences in age and pelvic dysfunction scores (obtained through pelvic floor questionnaires) between symptomatic patients and asymptomatic volunteers were statistically significant (p < 0.001). Differences in body mass index were not statistically significant (p = 0.895).

Qualitative image analysisStraining effort

On the single-slice sequences, straining effort was good in 38 cases (84.4%), suboptimal in 6 cases (13.3%) and no straining effort was seen in one case (2.3%). On the multi-slice sequences, straining effort was good in 29 cases (64.4%), suboptimal in 12 cases (26.7%), and absent in 4 cases (8.9%). There was a statistically significant difference between the single-slice and multi-slice sequence (p = 0.003) (Fig. 1).

Fig. 1

Bar chart depicting the differences in distribution and proportion of subjectively assessed straining effort in the single-slice and multi-slice sequences (p = 0.003)

Visibility of organ points

The organ points were always visible in the multi-slice sequence. For the single-slice sequence, the bladder point was seen in all cases (100%) at rest and during straining. At rest, the cervix point was not visible in 5 cases (11.1%), partly visible in 9 cases (20%) and fully visible in 31 cases (68.9%). At straining, the cervix point was not visible in 6 cases (13.3%), partly visible in 9 cases (20%) and fully visible in 30 cases (66.7%). At rest and during straining, the anorectal junction was partly visible in one case (2.2%) and completely visible in all other cases (97.8%) (Supplementary Fig. 1).

Quantitative analysisPelvic organ point measurements in asymptomatic volunteers

At rest, the mean positions of bladder, cervix, and anorectal junction were − 2.8 cm (± 0.2 cm), − 5.4 cm (± 0.8 cm), and − 2.8 cm (± 0.6 cm) in the single-slice sequences. Same organ positions were − 2.7 cm (± 0.3 cm), − 5.0 cm (± 0.7 cm), and − 2.7 cm (± 0.7 cm) in the multi-slice sequences. There was a statistically significant difference between the single-slice and multi-slice sequences in the cervix position (p = 0.003), but not for the position of the bladder (p = 0.322) and anorectal junction (p = 0.465) (Fig. 2).

Fig. 2

Box plots comparing organ positions in the y-axis at rest (top row) and straining (bottom row) between single-slice (white box) and multi-slice sequences (grey box) in the asymptomatic volunteers. Asterisks indicate statistical significance

At straining, the mean positions of bladder, cervix, and anorectal junction were − 1.1 cm (± 1.1 cm), − 3.2 cm (± 1.1 cm), and − 0.7 cm (± 1.4 cm) in the single-slice sequences. Same organ positions were − 1.5 cm (± 1.0 cm), − 3.4 cm (± 1.1 cm), and −1.0 cm (± 1.3 cm) in the multi-slice sequences. There was a statistically significant difference between the single-slice and multi-slice sequences in the bladder position (p = 0.019), but not for the position of the cervix (p = 0.223) and anorectal junction (p = 0.158) (Fig. 2).

At rest, the mean difference (and range) between measurements in the single- and multi-slice sequence was − 0.1 cm (− 0.8 to + 0.2 cm) for the bladder, − 0.4 cm (− 1.4 to + 0.4 cm) for the cervix, and − 0.1 cm (− 0.7 to + 0.7 cm) for the anorectal junction. At straining, the mean differences were + 0.4 cm (− 0.3 to + 2.5 cm) for the bladder, + 0.2 cm (− 0.9 to + 1.8 cm) for the cervix, and + 0.3 cm (− 0.7 to + 2.5 cm) for the anorectal junction. All but one measurement in each compartment were inside the 95% confidence intervals (CI).

Pelvic organ point measurements in symptomatic patients

At rest, the mean positions of bladder, cervix, and anorectal junction were − 2.0 cm (± 0.6 cm), − 3.8 cm (± 2.0 cm), and − 1.7 cm (± 0.8 cm) in the single-slice sequences. Same organ positions were − 2.1 cm (± 0.6 cm), − 3.7 cm (± 2.0 cm), and − 1.6 cm (± 0.9 cm) in the multi-slice sequences. There was no statistically significant difference between the single-slice and multi-slice sequences (bladder: p = 0.927, cervix: p = 0.274, anorectal junction: p = 0.280) (Fig. 3).

Fig. 3

Box plots comparing organ positions in the y-axis at rest (top row) and straining (bottom row) between single-slice (white box) and multi-slice sequences (grey box) in symptomatic patients. Asterisks indicate statistical significance

At straining, the mean positions of bladder, cervix, and anorectal junction were + 1.1 cm (± 1.8 cm), − 0.7 cm (± 2.9 cm), and + 0.7 cm (± 1.3 cm) in the single-slice sequences. Same organ positions were + 0.4 mm (± 1.7 cm), − 1.4 cm (± 2.6 cm), and + 0.4 cm (± 1.3 cm) in the multi-slice sequences. There was a statistically significant difference between the single-slice and multi-slice sequences for all three organ points (bladder: p < 0.001, cervix: p = 0.011, anorectal junction: p = 0.042) (Fig. 3).

At rest, the mean difference (and range) between measurements in the single- and multi-slice sequence was 0.1 cm (− 0.2 to + 1.1 cm) for the bladder, − 0.1 cm (− 1.6 to + 1.4 cm) for the cervix, and 0.0 cm (− 1.0 to + 1.2 cm) for the anorectal junction. At straining, the mean difference was + 0.7 cm (− 0.3 to + 3.4 cm) for the bladder, + 0.7 cm (− 0.8 to + 5.7 cm) for the cervix, and + 0.4 cm (− 0.9 to + 2.4 cm) for the anorectal junction. In each compartment, there were one to two measurements outside of the 95% CI (Figs. 4, 5, 6).

Fig. 4

Bland–Altman plots of the bladder organ point measurements on single-slice and multi-slice sequences in volunteers (left row) and patients (right row) at rest (top row) and straining (bottom row). The black line depicts the mean difference. The top and bottom grey lines denote the upper and lower limits of agreement

Fig. 5

Bland–Altman plots of the cervix organ point measurements on single-slice and multi-slice sequences in volunteers (left row) and patients (right row) at rest (top row) and straining (bottom row). The black line depicts the mean difference. The top and bottom grey lines denote the upper and lower limits of agreement

Fig. 6

Bland–Altman plots of the anorectal joint measurements on single-slice and multi-slice sequences in volunteers (left row) and patients (right row) at rest (top row) and straining (bottom row). The black line depicts the mean difference. The top and bottom grey lines denote the upper and lower limits of agreement

Interreader agreement

For the single-slice sequences at rest and straining, the interobserver agreement was excellent for the bladder (ICC 0.989, 95% CI 0.971–0.995) and cervix (ICC 0.953, 95% CI 0.880–0.981) and good for the anorectal junction (ICC 0.679, 95% CI 0.189–0.873).

For the multi-slice sequences at rest and straining, the interobserver agreement was excellent for the bladder (ICC 0.968, 95% CI 0.918–0.987), cervix (ICC 0.917, 95% CI 0.790–0.967), and anorectal junction (ICC 0.872, 95% CI 0.676–0.949).

Pelvic organ prolapse gradingAsymptomatic volunteers

Three cases (13.6%) had a grade 1 descent of the anterior compartment on the single-slice sequence, of which two were also evident on the multi-slice sequence and one was not. In that case, the bladder point was measured at 1.5 cm below the PCL in the single-slice sequence but only at 1.0 cm below PCL in the multi-slice sequence (difference of 0.5 cm), hence, the formal threshold of 1.0 cm for calling a grade 1 descent was not reached on the multi-slice sequence (Table 2).

Table 2 Cases with discrepancy of POP grading between measurements on single-slice and multi-slice sequences

No descent of the middle compartment was seen.

Two cases (9.0%) had a grade 1 descent of the posterior compartment on single-slice sequence, which was also evident on the multi-slice sequence.

Symptomatic patients

Fourteen symptomatic patients (60.8%) presented with a descent of the anterior compartment (grade 1: n = 6, grade 2: n = 8) on the single-slice sequence. In 7 of these 14 cases (50%), the multi-slice sequence did not depict a descent (n = 5) or only depicted a grade 1 instead of grade 2 descent (n = 2). Of those cases that were not depicted in the multi-slice sequence, three cases had a grade 1 descent and two cases a grade 2 descent. Three of those five missed cases had either a suboptimal (n = 2) or no visible (n = 1) straining effort in the multi-slice sequence. In those two cases where a grade 1 instead of grade 2 descent was depicted on the multi-slice sequence, the straining effort was scored as good and the difference in measurements between the single- and multi-slice sequence was 0.1 and 0.6 cm, respectively (Table 2).

Four symptomatic patients (17.3%) presented with a descent of the middle compartment (grade 1: n = 3, grade 2: n = 1) on the single-slice sequence. One of these cases with grade 1 descent was not captured on the multi-slice sequence despite a subjectively good straining effort (the difference in measurement between the single- and multi-slice sequence was 1.5 cm) (Table 2).

Seven symptomatic patients (30.4%) had a descent of the posterior compartment (grade 1: n = 6, grade 2: n = 1) on the single-slice sequence. The multi-slice sequence did not capture a descent in 3 of those cases. These included two cases with a grade 1 descent (one with suboptimal straining effort, the other with a good straining effort, but did not meet the threshold for a grade 1 descent with a difference of 0.5 cm between the single- and multi-slice sequence) and one case with a grade 2 descent, where there was no visible straining effort in the multi-slice sequence (Table 2).

Figure 7 shows an example of single-slice and multi-slice sequence images from a symptomatic patient with POP.

Fig. 7

38-year old primiparous woman with symptomatic pelvic organ prolapse in involving all three compartments and a POP-Q score of III. Sagittal single-slice (a, c) and multi-slice (b, d) MR images at rest (a, b) and maximum straining (c, d) are shown. The bladder (white arrow), cervix (black arrow) and anorectal junction (white arrowhead) are annotated. At rest, the bladder point was 1.6 cm and 1.6 cm, the cervix point was 4.1 cm and 3.7 cm, and the anorectal junction was 1.4 cm and 1.0 cm above the PICS line in the single-slice and multi-slice sequences, respectively. During maximum straining, the bladder point was 4.1 cm and 3.4 cm below, the cervix point was 0.2 cm and 0.0 cm above, and the anorectal junction was 3.1 cm and 3.1 cm below the PICS line in the single-slice and multi-slice sequence, respectively. Differences between single and multi-slice sequences were small (0.1–0.4 cm at rest and 0.0–0.7 cm at straining) and there was no difference in the POP grading between the single-slice and multi-slice sequence (grade 2 descent of the anterior compartment, and grade 1 descent of the posterior compartment)