Dynamic three-dimensional spiral computed tomographic cysto-urethrography: a novel technique for evaluating post-traumatic posterior urethral defects

Authors


T. Osman, Department of Urology, Ain-Shams University, Cairo, Egypt.
e-mail: tarekosman77@hotmail.com

Abstract

OBJECTIVE

To present a new method of identifying the anatomy of posterior urethral distraction defects (PUDDs) using three-dimensional spiral computed tomography/cysto-urethrography (CTCUG), as conventional two-dimensional CUG can give a false interpretation of the exact anatomy.

PATIENTS AND METHODS

Twenty-one patients presenting with a PUDD were assessed between February 2001 and October 2002. All patients initially underwent combined ascending and micturating CUG (ACUG), followed by CTCUG analysed using special software. In this technique all soft-tissue densities were subtracted from the volume of interest leaving only high-density images, i.e. pelvic bony structure and the contrast medium-filled bladder and urethra. The images were taken over a sequence and 36 different images viewed by ‘roll and spin’ techniques, each of which had a different plane of vision. Images were supplied as static CT films and as a movie on a compact disk using a computer program.

RESULTS

The technique allows one CT image to be viewed from 36 different angles both dynamically and statically, allowing the surgeon the unique opportunity to view the PUDD through several planes and precisely identify its anatomy. By comparing the data obtained with this technique to those obtained by conventional ACUG, and each in relation to the operative findings, the following aspects could be more thoroughly evaluated by CTCUG: the location of the distraction injury, the length of the distraction defect, the degree and direction of urethral end-alignment, the relation of the ectopic bony fragments and bone callus to the urethra, and the presence of various pathological defects, e.g. fistulae, false passages and diverticulae.

CONCLUSION

With CTCUG, both static and dynamic images can be obtained, allowing the easier staging of a PUDD and thus better surgical planning and consequently better results from reconstruction.

Abbreviations
PUDD

posterior urethral distraction defect

3D

three-dimensional

CTCUG

dynamic 3D spiral CT cysto-urethrography

ACUG

ascending/voiding cysto-urethrography.

INTRODUCTION

Blunt pelvic trauma results in posterior urethral distraction defects (PUDDs) in ≈ 10% of cases [1]; such injury commonly involves the membranous urethra at the point of departure from the bulbospongiosum, at the prostatomembranous junction, or at any point between its departure and the apex of the prostate [1]. PUDDs are complex pathologically, involving displacement and misalignment of the severed urethral ends with intervening and surrounding fibrosis. Detached bony fragments and callus formation add to the pathological complexity. For a successful repair of a PUDD it is necessary to identify the specific anatomy of the distraction defect before undertaking any treatment.

The classic approach for evaluating a PUDD is through cysto-urethrography (CUG), particularly while the patient is attempting to void. However, this study can often give a false interpretation of the exact anatomy of the distraction defect on many occasions. Here we present a new method of precisely identifying the anatomy of a PUDD, using dynamic three-dimensional (3D) spiral CT CUG (CTCUG).

PATIENTS AND METHODS

Twenty-one patients presenting with a PUDD were imaged using CTCUG between February 2001 and October 2002. All patients initially underwent conventional ascending and voiding CUG (ACUG) and on another day CTCUG. The study starts by filling the bladder through a suprapubic catheter with diluted contrast medium (normal saline, 2 : 1), with filling continued until the patient perceives a comfortably full bladder sensation. Contrast medium is then injected retrogradely through the urethra and scanning commenced, extending from the iliac crest above to the end of the urethra below, and followed by another scan with the same parameters while the patient is attempting to void. The technical parameters were uniform throughout, using 120 kV and 250 mA with a slice thickness of 10 mm and scan time of 1 s. Cross-sectional images were obtained, reviewed and the volume of interest chosen, and then the following steps added: (i) Cross-sectional images were grouped according to the examination phase and then transferred to a separate image processor where they were reconstructed in 3D using special software (Easy Vision® v4.3, Philips Medical Systems, the Netherlands); 1 mm slice thickness images were taken; (ii) All soft tissue densities were removed (subtracted) from the volume of interest, leaving only high-density images, i.e. the pelvic bony structures and the contrast medium-filled bladder and urethra; (iii) the 3D rendering process was further extended by adding colour to enhance the perception of anatomical details, where the pelvic bony structures were given one colour and the bladder and urethra another, then both added together on one image; (iv) these images were used to assess the bony pelvic structures and their relation with the bladder and urethra, together with any displaced bony fragments. The bone images were then subtracted to isolate any urethral pathology; (v) the images were taken over a film sequence, with 36 different images viewed by the ‘roll’ technique and another 36 by the ‘spin’ technique. Each of the 36 different images had a different plane of vision, and the images were supplied as a hardcopy and stored on a CD; (vi) the images were then processed in another program (ACDSee32) which allows the images to be viewed as a ‘movie’, and supplied both on CT film as usual, and as a CD for viewing by the surgeon. The average imaging time was 30 s and the average processing time 20 min.

RESULTS

Before each operation the surgeon reviewed the conventional CUG, the CTCUG images and the CD movie. The data obtained from each radiological examination were compared with the result disclosed at surgery. CTCUG was more informative than conventional radiology in several aspects; the location and the length of the distraction defect; the direction of alignment or misalignment; the bone anatomy (ectopic fragments, callus); and the presence of additional urinary pathology (fistulae, false passages, diverticulae).

The urethra was reconstructed through either a perineal approach or combined perineal and suprapubic approaches. The following examples show how CTCUG was more useful than the conventional radiological study.

For the location and length of the PUDD, Fig. 1a shows a conventional urethrogram from a 26-year-old man who presented with a PUDD 14 months after a motor car accident. The urethrogram showed a PUDD involving the whole prostatomembranous urethra, but neither the bladder neck nor the prostatic urethra were visualized. The distraction defect was apparently ≈ 4 cm long. The urethral end alignment was not evaluated but on CTCUG the bladder neck and the proximal part of the prostatic urethra were visualized, and the urethral ends were aligned, i.e. in the same plane (Fig. 1b). Moreover, when CTCUG also clearly showed the proximal prostatic urethral segment, particularly in the ‘spin’ sequence of images, the PUDD was found to be only 2 cm long, in contrast to the finding on conventional imaging. The operative findings confirmed the CTCUG data, and an adequate prostatic urethral segment was exposed and end-to-end anastomosis carried out successfully.

Figure 1.

A, Conventional ACUG of a patient with a PUDD showing the defect involving the whole prostatomembranous urethra, but with neither the bladder neck nor prostatic urethra visible. The PUDD appears to be > 4 cm long. b, CTCUG shows the bladder neck and proximal prostatic urethra; the distraction defect is only 2 cm long.

For the bony anatomy Fig. 2a shows a ACUG from a 21-year-old man who presented with a PUDD 8 months after a road traffic accident. Although the urethral stricture was evident the anatomy of the bone was not adequately identified and from this one film it was not possible to outline the amount and position of ectopic bone and callus in relation to the urethral ends. On CTCUG (Fig. 2b) and the series of ‘roll’ and ‘spin’ sequences, both static and dynamic, callus was found to be imposed on the urethra and an ectopic bone fragment from a fractured superior pubic ramus was apparent encroaching on the proximal membranous urethra. Initially a surgical repair was attempted through a perineal approach alone. The presence of extensive bone fragments necessitated a transpubic approach with pubectomy and excision of the ectopic bone. Had the findings of CTCUG been carefully considered before surgery it would probably have saved time and effort in trying to repair the PUDD through a perineal approach alone, and the combined approach would have been selected from the start.

Figure 2.

A, Conventional ACUG of a patient with a post-traumatic posterior urethral stricture showing the distorted posterior urethra and ill-defined bone anatomy. B, CTCUG shows callus imposing on the prostatic urethra and an ectopic bone fragment from a fractured superior pubic ramus encroaching on the proximal membranous urethra.

For identifying abnormal urinary pathology Fig. 3a shows a urethrogram from a 42-year-old man with a post-traumatic stricture and a probable posterior false passage. He had undergone repeated urethroscopic interventions before presenting to our institution. Diagnostic urethroscopy showed a patent urethra, and a dilated posterior urethra and bladder. However, despite these findings, the patient was obstructed and dribbled urine occasionally. CTCUG showed many false passages, from the previous interventions, that were not apparent on conventional urethrography (Fig. 3b,c).

Figure 3.

A, Conventional ACUG showing a post-traumatic stricture with a probable posterior false passage. B, CTCUG shows many false passages from previous endoscopic interventions, and C, the same image after bone subtraction.

Figure 4a is a urethrogram from a 27-year-old man presenting with a post-traumatic PUDD, showing a posterior urethral stricture and proximal extravasation of contrast medium. The CTCUG images (Fig. 4b,c) showed a clear lateral false passage extending to the bladder from a previous intervention. Both patients were explored through a perineal approach and the false passages excised, followed by urethroplasty.

Figure 4.

a, Conventional ACUG showing a poorly visualized posterior urethra with extravasation of contrast medium. b, CTCUG shows a well-defined lateral false passages and c, the bone-subtraction image.

DISCUSSION

Proper surgical planning is the cornerstone for the successful repair of a PUDD, and entails selecting the appropriate surgical approach, anticipating the need for bone or callus resection, with or without pubectomy, preparing for possible morbidity (e.g. blood transfusion, anaesthesia complications) and preoperative patient counselling (particularly about impotence and incontinence). Such surgical planning necessitates a thorough knowledge of the anatomy of the PUDD.

Important information required before intervention is the length of the urethral defect, the direction and degree of alignment of the prostato-membranous complex with the proximal bulbous urethra, the presence of fistulae and false passages, and identification of bone anatomy, particularly ectopic bone or callus from healed fractures. Conventional imaging can fail to fully characterise the PUDD; the overlap of bone fragments might hinder proper visualization of the urethra. Soft-tissue reaction near the PUDD can interfere with image quality and resolution. Often proper positioning of the patient with pelvic fractures is not possible. The techniques usually require a skilled radiologist and for the best results the urologist should be present, all of which may be difficult, costly and time-consuming. All these factors may result in improper surgical planning and consequently confusion during the surgical intervention.

MRI has been advocated for staging PUDDs [2,3] and using a body-coil can give useful information. Transaxial, coronal and sagittal views are required, and examining all three planes could be useful for evaluating the post-traumatic anatomy. Moreover, details about the corporeal tissue and the vasculature of the pelvis could be obtained, which might assist in determining the prognosis for erectile dysfunction [4]. However, MRI has several disadvantages for a PUDD; it is costly, no digital subtraction details of the urethra are obtainable, and it is unfamiliar to the urologist. Moreover, the benefit of MRI in evaluating potential erectile function is debatable as compromised erectile function is considered to be secondary to the trauma rather than reconstruction [5]. Consequently, we consider MRI to provide little prognostic information.

The present technique can be used to identify the anatomy of a PUDD, with one CT image viewed from 36 different angles as static and dynamic (movie) films. This feature allows the surgeon to view the PUDD through several planes and identify precisely its anatomy. Therefore, before treatment the condition can be classified primarily as a simple or complex surgical problem, the procedure easily planned and the patient adequately counselled. The technique might also be helpful for evaluating and planning early endoscopic re-alignment of post-traumatic posterior urethral defects, a procedure which has been encouraged recently by some. The main disadvantage of CTCUG is that it is relatively expensive, but it does not need an experienced operator and it is simpler than conventional radiology.

In conclusion, dynamic CTCUG is helpful for accurately evaluating a PUDD; considering the additional information that CTCUG may offer beyond conventional radiography, it may be a cost-effective tool. Nevertheless, further evaluation of the technique by different investigators is required.

Ancillary