DOI:10.2214/AJR.07.2149
AJR 2007; 189:1517-1522
© American Roentgen Ray Society
Management of Recurrent Urethral Strictures with Covered Retrievable Expandable Nitinol Stents: Long-Term Results
Eugene K. Choi1,2,
Ho-Young Song1,
Ji Hoon Shin1,
Jin-Oh Lim1,
Hyungkeun Park3 and
Choung-Soo Kim3
1 Department of Radiology and Research Institute of Radiology, University of
Ulsan College of Medicine, Asan Medical Center, 388-1 Poongnap-2 Dong,
Songpa-Gu, 138-736 Seoul, Korea.
2 Weill Medical College of Cornell University, New York, NY.
3 Department of Urology, University of Ulsan College of Medicine, Asan Medical
Center, Seoul, Republic of Korea.
Received October 4, 2006;
accepted after revision June 19, 2007.
Address correspondence to H. Y. Song.
Abstract
OBJECTIVE. The purpose of this study was to evaluate the long-term
clinical efficacy of temporary placement of covered retrievable stents in the
management of recurrent urethral strictures.
MATERIALS AND METHODS. During the period December
1998–December 2005, 32 men and one adolescent boy (mean age, 48.6 years;
range, 16–73 years) with recurrent urethral strictures underwent
fluoroscopically guided insertion of a total of 68 stents. Patients without
complications underwent elective stent removal 2 or 4 months after stent
insertion. Rates of clinical success (long-term clinical and radiographic
resolution of urethral strictures) were assessed. The Mann-Whitney U
test was used to compare the duration of stent placement in patients with
long-term clinical resolution with that in patients with stricture
relapse.
RESULTS. Clinical success was achieved in 18 (55%) of the 33
patients. The mean duration of stent placement in patients with clinical
success was significantly different from that in patients who had recurrences
(p < 0.0001). Stricture relapse did not occur in only four (20%)
of 20 cases of stent placement for 2 months. All 14 stent placements lasting
at least 4 months resulted in long-term resolution after a mean follow-up
period of 3.6 years. The most common complications necessitating early stent
removal were stent migration (33.8% of stents) and tissue hyperplasia (20.6%
of stents).
CONCLUSION. Placement of a covered retrievable stent for a minimum
of 4 months is effective in inducing long-term resolution of refractory
urethral strictures. Stent migration remains the largest obstacle in achieving
adequate duration of stent placement.
Keywords: fluoroscopy • interventional radiology • stent • urethral stricture
Introduction
Management of recurrent benign urethral strictures remains a therapeutic
challenge to urologists. Strictures managed with conventional treatments
(i.e., endoscopic urethrotomy, urethroplasty) often remain resistant to
therapy. Results of previous studies
[1,
2] indicate that 40–70%
of strictures managed with endoscopic urethrotomy recur within 2 years. Stent
placement is an evolving alternative for treatment of selected patients with
recurrent urethral strictures whose condition is unsuited to invasive
procedures. Despite promising early results with the use of a self-expanding
stent (UroLume, American Medical Systems)
[3,
4], recent reports
[5,
6] of long-term results have
been discouraging. The complication rates reported are as high as 55%, most of
the complications stemming from stent obstruction secondary to tissue
hyperplasia. These stents lack a covering material and thereby become easily
incorporated into the urethral wall with hyperplastic tissue ingrowth. Because
it is permanent, the stent must be removed surgically if complications
necessitate removal. Retrieval stents designed for temporary placement thus
have an advantage. Several types of temporarily placed stents
[7,
8] have been introduced. These
stents, however, are not suitable for the management of bulbomembranous
strictures that require stents that traverse the external sphincter. In
addition, the need for use of a rigid endoscopic forceps in stent removal
makes the procedure cumbersome.
The most recent design of retrievable stent was introduced by Song et al.
[9], who found promising
initial results in the treatment of 12 men with recurrent bulbar strictures.
Evaluation after a mean follow-up period of 18 months showed that 11 (92%) of
12 patients had been successfully treated with preservation of sphincter
function despite complete bridging of the external sphincter with the stent. A
follow-up study [10] showed
that this newly designed stent can be easily retrieved with a high technical
success rate and low complication rate. The purpose of our study was to assess
the long-term effectiveness of this stent in the management of recurrent
urethral strictures in 33 patients with a mean follow-up period of 3.6
years.
Materials and Methods
Patients
All patients provided informed consent to undergo urethral stent placement,
and the retrospective review was approved by our institutional review board.
During the period December 1998–December 2005, 70 retrievable expandable
nitinol stents were fluoroscopically placed in 35 consecutively enrolled male
patients with recurrent benign urethral strictures. Two patients were lost to
follow-up immediately after stent removal and were excluded. The final study
population included 32 men and one adolescent boy (mean age, 48.6 years;
range, 16–73 years) who had undergone placement of a total of 68 stents.
Seventeen patients received a single stent; seven patients, two stents; four
patients, three stents; three patients, four stents; one patient, six stents;
and one patient, seven stents. The causes of stricture included trauma in 31
cases (traffic accident, n = 10; fall, n = 14; and
indwelling Foley catheter, n = 7), surgery (prostatectomy for
prostate cancer) in one case, and an unknown factor in one case. Before stent
placement, all patients had undergone previous treatments that failed,
including one or more of visual internal urethrotomy, urethroplasty, and
balloon or sound dilation. These patients were referred to the interventional
radiology service by the urology service for fluoroscopic stent placement. The
stricture site was determined to be in the bulbomembranous region in all but
four patients. In those patients, the primary lesion was located in the penile
urethra in three cases and in the prostatic urethra in one case. The mean
length of the strictures managed with stent placement was 2.69 cm (range,
1–10 cm).
Stent Insertion and Removal
The stent has been described previously
[9]. In brief, the stent was
developed from a 0.1-mm-diameter nitinol wire filament in a tubular
configuration and was covered with either polyurethane (Chronoflex, Cardiotech
International) (n = 25) or polytetrafluoroethylene (AG Fluoropolymer)
(n = 43) to prevent mucosal hyperplasia through the stent wires
(Fig. 1). The stent also
featured drawstrings attached to the lower inner margin, allowing the stent to
be easily retrieved with a hook catheter. The stent was 10 mm in diameter when
fully expanded and 40–55 mm long. The stent and introducer set were made
by a local manufacturer (Taewoong).
Before stent placement, the site, severity, and length of the stricture
were evaluated with urethrography and uroflowmetry. Oral prophylactic
antibiotic therapy was begun 1 day before the procedure and continued for at
least 1 week. With the patient placed in the left anterior oblique position
with knees bent, disinfection of the external urethral orifice and anesthesia
of the urethra were achieved with 0.05% chlorhexidine and 10 mL of lubricating
jelly containing 2% lidocaine. Retrograde urethrography was performed with
fluoroscopic guidance with the sites of the stricture and the external
sphincter marked on the patient's skin with radiopaque markers. The skin
markings were made to estimate the stricture length and its location in
relation to the external sphincter.
A 0.035-inch guidewire (Radifocus M, Terumo) was inserted through the
urethra across the stricture into the urinary bladder under fluoroscopic
guidance. The stent was placed through a straight 5-French graduated sizing
catheter (Cook) to the proximal part of the stricture to measure the length of
the stricture. A 9-French sheath with a dilator was then passed over the
guidewire in the urethra and was advanced according to a technique described
previously [9]. Retrograde
urethrography was performed to verify the position of the stent.
On the basis of our experience of stenting benign strictures of the
esophagus [11], it was
initially our policy to electively remove the stent after 2 months of
placement. However, reports of frequent stricture recurrences in a previous
study [9] prompted extension of
the duration of stent placement to 4 months. The stent was removed under
fluoroscopic guidance with a 9-French retrieval set according to the technique
reported previously [9,
10].
Follow-Up and Analysis
In all patients, retrograde urethrography, uroflowmetry, estimation of
postvoiding residual urine volume, and urine cultures were performed 1 week
after stent insertion and then every 4 weeks until the stent was removed.
Follow-up studies with retrograde urethrography were performed 1, 3, and 12
months after removal of the stent to assess recurrence. When the findings on
urethrography showed full expansion of the stent, urethroscopy was performed
to verify the patency of the stent. Patients were questioned with regard to
the frequency and urgency of micturition and level of continence. All patients
were contacted by telephone every 6 months until the end of the study.
The rate of clinical success, defined as long-term clinical and
radiographic resolution of urethral strictures after stent removal without the
need for additional intervention (i.e., balloon dilation, urethrotomy) was
assessed. To determine whether duration of stent placement and primary
stricture length play a role in the achievement of long-term resolution, we
classified the patients into a group with recurrence and one without
recurrence. We used the Mann-Whitney U test to compare the groups
with regard to mean duration of stent placement and stricture length. Clinical
success rates were analyzed according to location of urethral stricture. A
two-sided p < 0.5 was considered to indicate statistical
significance. All statistical analysis was performed with the SPSS program
version 11.5.
Results
Stent Placement and Removal
All patients voided well after stent placement. Whereas the mean maximum
urine flow rate was 5.0 mL/s (range, 3–7 mL/s) before stent placement,
the flow rate 1 week after stent placement was 27.1 mL/s (range, 16–40
mL/s). All patients reported mild urgency and discomfort at the site of stent
placement. These problems resolved spontaneously within 1 week after stent
placement. Fifteen patients reported minor postmicturition dribbling and nine
patients reported difficulty with continence, but spontaneous resolution of
these symptoms occurred within 4 weeks of stent placement. Mild hematuria
occurred after removal of seven stents, but bleeding resolved spontaneously
within 4 hours in these patients.
A total of 68 stents were removed from a total of 33 patients
(Table 1). Of the 68 stents
placed, only seven stents were electively removed after 2 months, and 20 were
removed after 4 months. The other 41 stents were prematurely removed because
of complications. The most common complication necessitating premature
retrieval was stent migration, occurring in the cases of 23 (33.8%) of 68
stents. In three patients, three stents migrated proximally into the urinary
bladder and were removed via cystostomy. Tissue hyperplasia of the urothelial
epithelium resulting in luminal narrowing at the proximal or distal ends of
the stent or both and concomitant obstructive symptoms were the primary
reasons for premature removal of 14 (20.6%) of the 68 stents. Other
complications resulting in premature stent removal included stent deformity
(n = 3) and severe pain (n = 1). In three patients (patients
6, 8, and 20 [Table 1]),
additional stents were inserted coaxially to overlap one or both ends of the
primary stent to stem the growth of hyperplastic tissue (Figs.
2A,
2B,
2C,
2D, and
2E) to allow extension of the
duration of stent placement to at least 16 weeks, resulting in long-term
resolution of the stricture.
View larger version (147K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2A —16-year-old boy with history of straddle injury presented
with recurrent 2-cm stricture in bulbar urethra. Left anterior oblique
retrograde urethrogram shows site of stricture (arrow) before stent
placement.
|
|
View larger version (129K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B —16-year-old boy with history of straddle injury presented
with recurrent 2-cm stricture in bulbar urethra. Left anterior oblique
retrograde urethrogram at stent placement shows 4-cm covered retrievable stent
(arrow) in bulbar urethra.
|
|
View larger version (138K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2C —16-year-old boy with history of straddle injury presented
with recurrent 2-cm stricture in bulbar urethra. Left anterior oblique
retrograde urethrogram 7 weeks after B shows early tissue hyperplasia
at distal end of stent (arrow).
|
|
View larger version (122K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2D —16-year-old boy with history of straddle injury presented
with recurrent 2-cm stricture in bulbar urethra. Left anterior oblique
retrograde urethrogram shows insertion of overlapping second stent
(arrow) at distal end of initial stent (arrowhead). Second
stent extended duration of first stent placement to 4 months.
|
|
View larger version (132K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2E —16-year-old boy with history of straddle injury presented
with recurrent 2-cm stricture in bulbar urethra. Left anterior oblique
retrograde urethrogram after stent removal shows good passage of contrast
material through former stricture site. Patient has been symptom free over 6
years of follow-up.
|
|
Follow-Up and Analysis
Patients were followed up for a mean of 174 weeks (range, 8.4–374
weeks) after the last stent removal. Clinical success was achieved by 18 (55%)
of the 33 patients. Four of the patients had a stent in place for 2 months,
and 14 patients had a stent in place for of 4 months
(Table 1). The mean duration of
stent placement in patients with clinical success (15.8 ± 4.57 weeks;
range, 8–22 weeks) was significantly different from that of patients
with recurrence (6.46 ± 5.09 weeks; range, 1–11 weeks) (p <
0.0001). Only four (20%) of 20 stents in place for 2 months did not result in
stricture relapse. All 14 stents in place for at least 4 months resulted in
long-term resolution (Figs. 3A,
3B,
3C, and
3D). One patient underwent six
1- to 3-month stent placements because of stricture relapse and complications
necessitating early stent removal. This patient achieved long-standing
resolution after placement of a seventh stent for 17 weeks.
View larger version (128K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3A —60-year-old man with history of traumatic stricture of bulbar
urethra underwent stent placement after three failed trials of optical
urethrotomy. Left anterior oblique retrograde urethrogram shows site of bulbar
stricture (arrow) before stent placement.
|
|
View larger version (130K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3B —60-year-old man with history of traumatic stricture of bulbar
urethra underwent stent placement after three failed trials of optical
urethrotomy. Left anterior oblique retrograde urethrogram shows 5-cm-long
covered retrievable stent (arrowheads) traversing external
sphincter.
|
|
View larger version (126K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3C —60-year-old man with history of traumatic stricture of bulbar
urethra underwent stent placement after three failed trials of optical
urethrotomy. Left anterior oblique retrograde urethrogram 4 months after stent
placement shows maintained patency of stent despite tissue overgrowth at
distal stent end (arrow).
|
|
View larger version (126K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3D —60-year-old man with history of traumatic stricture of bulbar
urethra underwent stent placement after three failed trials of optical
urethrotomy. Left anterior oblique retrograde urethrogram obtained 3 years
after C shows well-maintained patency at former stricture site. Patient
has not had relapse after 4 years of follow-up.
|
|
The mean stricture length in the patients successfully treated (2.2
± 0.08 cm) was not statistically significantly different from that of
patients who experienced relapse (3.38 ± 2.25 cm) (p =
0.0715). With regard to stricture location, 15 (52%) of 29 patients with
bulbomembranous strictures and two (67%) of three patients with penile
strictures had long-term resolution with stent placement. The one patient with
a prostatic stricture also had successful long-term resolution.
Discussion
There is a high predilection for recurrence of urethral strictures with the
conventional treatments of endoscopic urethrotomy and dilation. The ideal
management of recurrent strictures, therefore, would entail a noninvasive
procedure that would result in long-term elimination of the obstruction with
minimal complications. The stent would also have to be well tolerated at and
around the external sphincter, especially given the resultant complications of
incontinence in cases of stent-induced injury to the sphincter
[3–6].
The results of this study indicate that placement of a retrievable covered
expandable stent may be such an option as long as stent placement of long-term
duration (> 4 months) is achieved. We hasten to point out that multiple
trials of stent placement do not appear to be additive with regard to
treatment; that is, continuous stent placement with a duration of 4 months is
not equivalent to four separate trials of stent placement of 1 month. This
issue is exemplified by the patient who achieved long-term resolution only
after achieving a stent placement duration of 4 months after previously
undergoing six separate trials lasting 1–3 months. This finding suggests
that long-term urethral stenting achieves stricture resolution in a
stricture-remodeling process that if not allowed sufficient time to proceed to
completion results in complete stricture relapse to the initial state. For a
minority of patients, 2 months of stenting was sufficient. A majority of
patients, however, needed at least 4 months of stent placement irrespective of
the number of previous trials of stent placement.
A complication of prolonged stent placement is induction of luminal
narrowing at the stent ends. Tissue hyperplasia resulting in recurrence of
obstructive symptoms was a major complication resulting in premature stent
removal and eventual recurrence of the primary stricture. We were able to
circumvent this complication in three cases by inserting additional stents at
one or both ends of the primary stent, thereby achieving the sufficient
duration of stent placement that is essential for long-term resolution.
Although addition of multiple stents can induce new sites of tissue
hyperplasia, resolution of stent-induced hyperplasia after stent removal, as
found in a previous study
[12], should not make this a
major concern. Moreover, it has been reported that patients who had needed
multiple stents for a variety of reasons (i.e., recurrent stricture adjacent
to stent, stent migration, underestimation of stricture length, tissue
hyperplasia within and at the ends of the stent) achieved the same efficacious
and durable results as reported for patients with single stents
[13].
We emphasize that the use of covering material and ease of retrievability
are important features integral to the effectiveness of a stent. The
polyurethane or polytetrafluoroethylene covering material minimizes the risk
of tissue ingrowth through the stent lumen. In comparison, the widely used
bare UroLume stent is prone to ingrowth of tissue hyperplasia through the
stent wiring, necessitating surgical intervention for removal. With the
retrievable feature of our stent, we can insert a stent temporarily throughout
stricture remodeling, after which the stent is removed before the occurrence
of further complications (e.g., tissue hyperplasia at the stent ends). As an
alternative, we can aggressively insert multiple stents to thwart hyperplastic
tissue growth and then subsequently remove all stents after an optimal
duration of stent placement is attained. Drug-eluting stents are under
investigation to reduce tissue hyperplasia associated with stent placement and
may soon obviate use of multiple stents
[14]. In patients in whom
stent placement is not well-tolerated, radio-nuclide-filled balloon dilation
has been used with success in preliminary studies
[15,
16].
The most common complication necessitating premature stent removal is stent
migration, a known problem associated with covered stents. We have routinely
used stents 10 mm in diameter. The optimal stent diameter, however, is
undetermined. Using stents of a wider diameter in patients with problems with
migration may help anchor the stent, and we are exploring this strategy.
An important additional advantage of the stent used in this study is
preservation of external sphincter function after stent removal. Although the
stent achieves sufficient radial force to dilate the stricture, its soft and
flexible characteristics minimize the risk of permanent damage to the external
sphincter. As a result, despite initial difficulties with continence in
association with stent placement, complete resolution eventually occurred in
all patients. Therefore, we propose an expansion of the indications for stent
placement to include stricture near the external sphincter.
Despite a limited number of patients with penile strictures treated with
stent placement, our findings suggest a potential role for use of stents in
this highly mobile part of the urethra that has been conventionally treated
with progressive dilations or optical urethrotomy
[7]. None of the patients with
penile strictures reported interference with erections or sexual activity.
Our long-term results indicate that temporary stent placement should be
considered the treatment of choice in cases of recurrent urethral stricture.
We have used stent placement only in patients who have undergone at least one
unsuccessful trial of conventional treatment, including visualized internal
urethrotomy and urethroplasty. Further investigation should involve an
examination of the efficacy of stent placement as the first-line treatment of
patients with urethral strictures.
Temporary placement of a covered retrievable stent for an extended duration
is effective in inducing long-term resolution of refractory urethral
strictures. The stent is well-tolerated in the bulbomembranous urethra,
including the region surrounding the external sphincter, with minimal
long-term complications. Stent migration, however, remains the largest
obstacle to achieving the adequate duration of stent placement critical for
achieving long-term resolution.
References
- Holm-Nielsen A, Schultz A, Moller-Pedersen V. Direct vision
internal urethrotomy: a critical review of 365 operations. Br J
Urol 1984; 56:308
-312[Medline]
- Pitkamaki KK, Tammela TL, Kontturi MJ. Recurrence of urethral
stricture and late results after optical urethrotomy: comparison of strictures
caused by toxic latex catheters and other causes. Scand J Urol
Nephrol 1992; 26:327
-331[Medline]
- Milroy E, Allen A. Long-term results of UroLume urethral stent for
recurrent urethral strictures. J Urol1996; 155:904
-908[CrossRef][Medline]
- Sertcelik N, Sagnak L, Imamoglu A, Temel M, Tuygun C. The use of
self-expanding metallic urethral stents in the treatment of recurrent bulbar
urethral strictures: long-term results. BJU Int2000; 86:686
-689[CrossRef][Medline]
- Hussain M, Greenwell TJ, Shah J, Mundy A. Long-term results of a
self-expanding wallstent in the treatment of urethral stricture.
BJU Int 2004; 94:1037
-1039[CrossRef][Medline]
- De Vocht TF, van Venrooij GE, Boon TA. Self-expanding stent
insertion for urethral strictures: a 10-year follow-up. BJU
Int 2003; 91:627
-630[CrossRef][Medline]
- Yachia D, Beyar M. New treatment modality for penile urethral
strictures using a self-expanding and self-retaining coil stent, UroCoil:
follow-up of 16 months after removal of the stent. Eur
Urol 1993; 24:500
-504[Medline]
- Yachia D, Beyar M. New, self-expanding, self-retaining temporary
coil stent for recurrent urethral strictures near the external sphincter.
Br J Urol 1993;71
: 317-321[Medline]
- Song HY, Park H, Suh TS, et al. Recurrent traumatic urethral
strictures near the external sphincter: treatment with a covered, retrievable,
expandable nitinol stent—initial results.
Radiology 2003;226
: 433-440[Abstract/Free Full Text]
- Shin JH, Song HY, Park H, et al. Removal of retrievable
self-expandable urethral stents: experience in 58 stents. Eur
Radiol 2006; 16:2037
-2043[CrossRef][Medline]
- Song HY, Jung HY, Park SI, et al. Covered retrievable expandable
nitinol stents in patients with benign esophageal strictures: initial
experience. Radiology 2000;217
: 551-557[Abstract/Free Full Text]
- Ko GY, Kim GC, Seo TS, et al. Covered, retrievable, expandable
urethral nitinol stent: feasibility study in dogs.
Radiology 2002;223
: 83-90[Abstract/Free Full Text]
- Tillem SM, Press SM, Badlani GH. Use of multiple UroLume
endourethral prostheses in complex bulbar urethral strictures. J
Urol 1997; 157:1665
-1668[CrossRef][Medline]
- Shin JH, Song HY, Choi CG, et al. Tissue hyperplasia: influence of
a paclitaxel-eluting covered stent—preliminary study in a canine
urethral model. Radiology 2005;234
: 438-444[Abstract/Free Full Text]
- Shin JH, Song HY, Moon DH, et al. Effects of irradiation using a
radioisotope-filled balloon on tissue hyperplasia caused by stent placement in
a canine urethral model. Acta Radiol2006; 47:436
-443[CrossRef][Medline]
- Shin JH, Song HY, Moon DH, et al. Reduction of tissue hyperplasia
with a rhenium-188 mercaptoacetyltriglycine-3-filled balloon: preliminary
study in a canine urethral model. J Vasc Interv Radiol2004; 15:737
-743[Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
