Upright 0.5 T open MR defaecating proctography proof of concept study: the inter- and intra-rater variability of pelvic floor measures using seated proctography

We tested the feasibility, data quality, and reliability of an upright magnetic resonance defaecating proctography (uMRDP) technique using an Open 0.5 T ASG MRI scanner. Eight healthy volunteers (2 males) performed seated defaecation on a purpose-built radio-frequency commode coil in an Open scanner. An optimised T2-weighted HASTE sequence captured dynamic changes during all three phases of the Kegel manoeuvre. Inter- and intra-rater variability was measured from the pelvic floor (PF) metrices (M-line, H-line, distances of landmarks) extracted by two radiologists. All relevant PF landmarks could be identified and metrices were extracted with acceptable inter- and intra-rater variability. Intra-rater variation was marginal, with relative absolute differences ranging from 5 to 21% and 3.2-44%. Inter-rater variability was reported using correlation and Bland-Altman plots. Correlation between raters was satisfactory, with r2 > 0.93, and bias ranged from - 1.8 to 0.65 mm. Moreover, the limit of agreement in the Bland-Altman plot ranged from 5.8 to 20.4 mm, indicating satisfactory precision. The proposed uMRDP technique can be used as a feasible and reliable alternative to supine MRDP, without the necessity of gadolinium injection and bowel preparation. It can capture defaecation in a regular seated posture and can provide information complementary to standard-of-care fluoroscopic proctography for clinicians.