Imaging evaluation of female pelvic anatomy and pathology has been dominated in recent years by the continued advances in computed tomography (CT), MRI, and ultrasound imaging. No single modality is a one-size-fits-all for pelvic evaluation, and the various distinct tissue characteristics of female pelvic components are best seen by the technique that exploits their inherent contrast to best effect. CT has rapidly evolved with the development of slip-ring technology, and the latest developments in multidetector row technology, together with three-dimensional volume-rendering tools, represent the state-of-the-art in CT imaging of the pelvis.

Simply put, multidetector row CT represents an advance in CT technology that allows faster scanning time without compromise of resolution.¹ Scanning times are reduced by a number of factors, including faster gantry rotation (500 ms), multiple rows of detectors, and faster processing speeds. Most multidetector row scanners in clinical practice at present are at least eight times faster than their single-detector predecessors. These scanners still use the helical principle of continuous gantry rotation during patient translation allowing for volume acquisition of data. Detector design does vary between vendors and the two main categories are fixed and adaptive array designs, which determine the permutations of detector sizes available for imaging protocols. These newer detectors also have the property of increased heat loading capacity permitting long Z-axis coverage without interruption and multiphase imaging without interscan delay, when required. Slice widths of 0.5 mm can be routinely obtained, but in practice we rarely image smaller than 1 mm. The x-ray beam may be collimated to 1 mm for high-resolution studies such as pelvic musculoskeletal or pelvic vascular examinations, whereas for soft tissues or organs less noisy, 2.5-mm collimation will usually suffice. The whole arena of multidetector row CT is somewhat of a moving target, and changes in the number of detector rows and gantry rotation times defy any single definition of a universal standard for the present. Nonionic iodinated contrast agents are helpful in discriminating the pelvic vasculature and lymph nodes. Delayed contrast-enhanced imaging is used to improve bladder or ureter visualization and can be of value for pelvic vein thrombosis studies. Oral contrast is routinely administered to delineate bowel, and a 45-minute delay is required to properly opacify distal small bowel and colorectal tissues. Air may be insufflated rectally if a virtual colorectal examination is contemplated, and occasionally rectal administration of positive contrast can help define the structures adjacent to the perirectal space.

3D Volume Rendering

Figure 1. Left lateral volume-rendered view of the pelvis illustrating the relationship of the bladder (short solid arrow), uterus (long solid arrow), and rectum (open arrow). Courtesy of Leo P. Lawler, MD.

CT interpretation remains dominated by axial planar two-dimensional images. This is how most people were trained and indeed it suffices in most cases. However, the complex anatomy of the bone and soft-tissue structures of the female pelvis do not easily conform to such limited planar representation. With the aforementioned high-quality multidetector row CT studies now available, we can produce isotropic and near-isotropic data sets. This enables the potential to look at the CT study in an infinite number of planes and projections and yet maintain image resolution and quality (Figure 1). Volume rendering is the latest technique that can harness the three-dimensional information contained within a routine CT examination. Volume-rendering techniques take all the raw data density information and use them to simulate three-dimensional images that are of high fidelity to the originally acquired data set.² The density values may subsequently be manipulated through trapezoids that allow variation of opacity, brightness, width, and level to confer on the visualized structures a depth and perspective. Color-coding may be assigned when required. Volume-rendered images are produced by postprocessing at a workstation and current computer power allows real-time slab clip plane editing of data and image rotation. Thus, after a CT study is performed, images are tailored to the individual patient's anatomy so that normal and abnormal features may be shown to best effect.

Clinical Application

Figure 2. Anterior coronal volume-rendered view of the bony pelvis. Courtesy of Leo P. Lawler, MD.

Multidetector row CT can be applied anywhere its predecessors' sequential CT and single detector helical CT were used. It does all the things the previous scanners did, but faster and with improved image quality. Similarly, 3D volume rendering may be applied anywhere 2D planar imaging is applied. It is not a matter of choosing between 2D and 3D as current software allows easy transition between both, and 3D images now may be edited with minimal labor intensity. 3D volume rendering should rather be viewed as a supplemental tool one can apply to an image to potentially gain greater insight into the pelvic anatomy or pathology, somewhat akin to changing window settings. Benefit has been proved in pelvic musculoskeletal studies where nonaxial anatomic features such as the acetabulum or femoral head can be better displayed^3,4 (Figure 2). Although 2D imaging may adequately define a bony feature, its inter-relationship to other bony features is better appreciated on a volume image of all relevant structures. 3D volume rendering is of proven benefit in interpretation of CT angiography studies.^5,6 The pelvic arteries are increasingly important in planning interventional studies, and 3D volume renderings are key to assessing vessel tortuosity and stenoses for aortic stent placement (Figure 3). For the primary diagnosis and the staging of neoplasms specific to the female pelvis, CT is of great value in defining the primary mass and the direct or lymphatic extension. Volume rendering may aid in such cases in better defining the tumor margins and the pelvic spaces involved.

The Time Factor

Figure 3. Left anterior oblique projection of the pelvis demonstrating the iliac and femoral vasculature coursing through it. Courtesy of Leo P. Lawler, MD.

When a clinician orders a 3D study, the requisition must state that this is desired. The test is then coded for the conventional 2D study as well as a separate 3D code/charge (6070). The 2D data acquired is used to perform the 3D study, so no extra acquisition time is used. Although higher resolution protocols are utilized, the speed of contemporary scanners permits this without any significant time penalty.

Interpretation and printing of 3D images require extra time. Current software allows rendering and printing in real time without large editing procedures so that an experienced user can perform these tasks in 10 to 15 minutes depending on the complexity of the study and information required. Most software packages now have very user-friendly interfaces and do not assume a high level of prior computer experience. A 5-day course and routine application of the techniques for 1 month give a level of proficiency for most clinical use. The basic principles of use can subsequently be applied across a range of applications. Depending on the institution, the 3D rendering is performed by the radiologist or a technologist and is performed at the time of the 2D study or at a separate time.

Conclusion

The standard of care in CT pelvic imaging is high-quality, rapid volume data acquisition. Multidetector row CT represents the latest step in the evolution of such imaging techniques. One means to harness more of the potential of such data sets is the application of 3D volume-rendering techniques, which are now widely available and practical to use. Imaging of many of the inherent complex anatomic features of the female pelvis may benefit from application of these developments.

Leo P. Lawler, MD is assistant professor, The Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins Medical Institutions, Baltimore.

References:

1. Hu H, He HD, Foley WD, Fox SH. Four multidetector-row helical CT: image quality and volume coverage speed. Radiology. 2000;215:55-62.
2. Calhoun PS, Kuszyk BS, Heath DG, Carley JC, Fishman EK. Three-dimensional volume rendering of spiral CT data: theory and method. Radiographics. 1999;19:745-764.
3. Buckwalter KA, Rydberg J, Kopecky KK, Crow K, Yang EL. Musculoskeletal imaging with multislice CT. Am J Roentgenol. 2001;176:979-986.
4. Guy RL, Butler-Manuel PA, Holder P, Brueton RN. The role of 3D CT in the assessment of acetabular fractures. Br J Radiol. 1992;65:384-9.
5. Prokop M. Multi-slice CT angiography. Eur J Radiol. 2000;36:86-96.
6. Rubin GD, Dake MD, Semba CP. Current status of three-dimensional spiral CT scanning for imaging the vasculature. Radiol Clin North Am. 1995;33:51-70.