Three-dimensional pelvimetry by computed tomography

Area of Science:

• Radiology and Medical Imaging
• Obstetrics and Gynecology
• Anatomy

Background:

• Accurate assessment of pelvic dimensions is crucial for understanding childbirth complications.
• Computed tomography (CT) offers detailed anatomical visualization, but its utility in pelvimetry requires evaluation.
• Different postprocessing techniques may influence the reliability of CT-based pelvic measurements.

Purpose of the Study:

• To evaluate the agreement and variability of computed tomography (CT) pelvimetry using various postprocessing methods.
• To determine the optimal CT postprocessing technique for reliable pelvic dimension measurements.

Main Methods:

• Retrospective analysis of CT data from 25 patients.
• Six pelvimetric measurements were performed by two independent observers using four different postprocessing techniques: biplanar topograms, 1-mm slice multiplanar reconstructions, and 3D volume-rendered images from 1-mm and 5-mm slices.
• Interobserver agreement and variability were assessed using Bland-Altman analysis; a scanned human skeleton served as a reference.

Main Results:

• Three-dimensional (3D) volume-rendered images from 1-mm slices demonstrated the highest interobserver agreement (correlation coefficient 0.98).
• Excellent interobserver agreement (0.96-0.99) was observed for sagittal outlet and midpelvic diameter, transverse inlet diameter, and obstetric conjugate.
• CT and ruler measurements of a reference skeleton showed excellent agreement, validating the CT measurement accuracy.

Conclusions:

• Three-dimensional (3D) volume-rendered CT reconstructions provide pelvimetry with low interobserver variability.
• CT pelvimetry is a reliable method for obtaining precise pelvic anatomical information.
• This technique is suitable for identifying parameters relevant to dystocia in retrospective studies.