- Email: [email protected]
THE CURRENT ROLE OF THE ARTIFICIAL URINARY SPHINCTER FOR THE TREATMENT OF URINARY INCONTINENCE
0022-5347/05/1742-0418/0 THE JOURNAL OF UROLOGY® Copyright © 2005 by AMERICAN UROLOGICAL ASSOCIATION
Vol. 174, 418 – 424, August 2005 Printed in U.S.A.
DOI: 10.1097/01.ju.0000165345.11199.98
Review Articles THE CURRENT ROLE OF THE ARTIFICIAL URINARY SPHINCTER FOR THE TREATMENT OF URINARY INCONTINENCE MAHREEN HUSSAIN,* TAMSIN J. GREENWELL, SUZIE N. VENN
AND
ANTHONY R. MUNDY
From the Institute of Urology and Nephrology, University College London, London, United Kingdom
ABSTRACT
Purpose: The introduction of the artificial urinary sphincter (AUS) in 1972 was heralded as a revolution for the treatment of genuine stress incontinence. Initial enthusiasm was tempered by disappointment as complications occurred. The device has now been in routine clinical use for more than 30 years, and the indications and surgical principles involved in its use along with short-term and long-term outcomes are more clearly defined. Hence, we reviewed the literature to clarify the role of the AUS and offer a possible solution to its problems in the guise of a new sphincter. Materials and Methods: A MEDLINE search was performed and all articles relating to the role of the AUS for the treatment of urinary incontinence were reviewed. Results: The AMS 800 (American Medical Systems, Minnetonka, Minnesota) provides urinary continence in 73% of cases (range 61% to 96%) and it has a complication rate of 12% (range 3% to 33%) for mechanical failure, 4.5% to 67% for early infection/erosion, 15% for late erosion and 7% for delayed recurrent incontinence. The literature supports the role of the AUS as an important and reliable treatment modality for stress urinary incontinence and intrinsic sphincter deficiency. However, it is not suitable in all patients and its use for the management of hypermobility is controversial. Hence, careful patient selection according to indication is required with full preoperative counseling. Conclusions: Despite its reliability for achieving urinary continence the AMS 800 is not perfect. Newer devices, such as that being developed at our institution, may offer improved outcomes and decreased complication rates. KEY WORDS: urethra; urinary incontinence, stress; urinary sphincter, artificial; incontinence; complications HISTORY
The first description of an artificial urinary sphincter (AUS) in 1947 is credited to Foley,1 who developed an inflatable cuff wrapped around the penile urethra in a tunnel lined by skin and controlled by a detachable pump carried in the pocket. Scott implanted the first AUS in current use. The initial model, the AMS 721 (American Medical Systems), was introduced in 1972 (fig. 1).2 Valves controlled the direction of flow in the system and the pressure within it. Developments in design resulted in several changes, of which the most significant was the use of a balloon instead of valves to regulate pressure. Later models also saw the introduction of an entirely silicone cuff instead of a Dacron® reinforced silicone rubber cuff. A decrease in the number of components and connections culminated in the current 3-part AMS 800, which incorporates high quality aerospace components as a consequence of a collaboration between AMS and the National Aeronautics and Space Administration (fig. 2). First introduced in 1982, this has been the only AUS on the market for the last 20 years and more than 20,000 have been implanted.
FIG. 1. Original AMS 721 prosthetic urinary sphincter was first one to be implanted. It consisted of 4 main parts, namely reservoir, inflatable cuff and 2 pumping mechanisms for inflation and deflation, respectively.2
Submitted for publication June 2, 2004. * Correspondence: Institute of Urology and Nephrology, Middlesex INDICATIONS Hospital, 48 Riding House St., London W1W7EY, United Kingdom Our main indications for AUS insertion are post(telephone: 0044 207 679 9112; FAX: 0044 207 679 9584; e-mail: [email protected]). prostatectomy incontinence (PPI), sphincter weakness incon418
CURRENT ROLE OF ARTIFICIAL URINARY SPHINCTER FOR INCONTINENCE
419
AUS insertion. They require urodynamic and renal function assessment before and after implantation. Similar risks occur in patients with decreased compliance and small functional bladder capacity.5 The 2 groups should be treated with anticholinergic medication and, if this fails, augmentation cystoplasty should proceed6 to decrease contractility and increase compliance and functional bladder capacity simultaneously or before AUS insertion. CONTRAINDICATIONS
An absolute contraindication to the AUS is patient inability to manipulate the pump. AUS implantation should be avoided in patients at high risk for complications, eg women who have received radiotherapy to the pelvis. In our series the failure rate in this group was 100%.7 The AUS can be implanted around the bulbar urethra in men following radiotherapy to the pelvis with reasonable results, although the complication rate is higher. Previous bladder neck trauma, such as urethral injury in women, also carries a high risk of erosion. THE DEVICE
FIG. 2. Three-part AMS 800
Indications for AUS insertion3, 4 Dysfunction Congenital neurogenic: Spina bifida Sacral agenesis Shy-Drager syndrome Acquired neurogenic (spinal cord injury, multiple sclerosis) Congenital nonneurogenic (exstrophy-epispadias complex, cloaca, urogenital sinus complex) Acquired nonneurogenic: PPI Failed female anti-incontinence surgery Pelvic fracture
% Pts ⫾ 42 ⫾ 40 11 5 42 ⫾ 25 ⫾ 10 ⫾8
The AUS consists of 3 basic components, namely a pressure regulating balloon, an inflatable cuff and a control pump. The entire system is fluid filled and it functions hydraulically. Pressure in the system and, therefore, the occlusive force of the cuff is determined by the pressure regulating balloon. The balloon is made of a silicone polymer manufactured by a dip coating process,8 rendering it resistant to wear and tear. The pressure produced depends on the elasticity in the balloon wall, which in turn depends on the thickness of the wall, and the amount of fluid within it. The pressurevolume relationship is linear until it attains a plateau at between 16 and 24 ml of fluid, where it remains predictable within a narrow range (fig. 3).4 Pressure in the system is maintained except when the pump is squeezed. The choice of pressure is made by the surgeon intraoperatively. The ranges are 51 to 60, 61 to 70 and 71 to 80 cm H2O, while 41 to 50 and 81 to 90 cm H2O pressure cuffs are also available on special request. Cuffs are available in various sizes. All are 2 cm wide but they range from 4.0 to 11 cm long, allowing correct sizing of the bulbar urethra or bladder neck. To aid assembly the tubing is color coded. Tubing connecting the cuff to the pump is white and that connecting the balloon to the pump is black.
tinence due to neurogenic bladder dysfunction, intrinsic sphincter deficiency (ISD) and rare congenital causes of incontinence (see table).3, 4 PATIENT SELECTION AND ASSESSMENT
Full patient evaluation is essential prior to AUS insertion. It is important to establish that incontinence is sufficiently disabling to warrant surgical intervention. Urine must be sent for microscopy, culture and sensitivity, and sterility must be ensured prior to implantation. If cutaneous manifestations of incontinence such as ammonia dermatitis are present, a catheter should be placed for 2 weeks preoperatively to ensure intact skin with normal bacteriological flora at the time of sphincter implantation.4 It should then be removed shortly before surgery to decrease the risk of bacterial contamination by this route. Patient pre-assessment should be tailored to the indication. In those with PPI cystoscopy and urodynamics exclude coexisting bladder outlet obstruction or detrusor overactivity. If present, they must be treated prior to implanting an AUS. Those with neurogenic bladder dysfunction who have low volume detrusor hyperreflexia may be at risk for upper tract damage due to higher detrusor contractile pressure following
FIG. 3. New sphincter
420
CURRENT ROLE OF ARTIFICIAL URINARY SPHINCTER FOR INCONTINENCE DEACTIVATION AND ACTIVATION
At the end of the implantation procedure the device is deactivated. The control pump incorporates a valve, a refill delay resistor and a deactivation button. When this button is pressed, it stops fluid from being transferred among the 3 components, allowing the surgeon to maintain the sphincter cuff in a deflated state for a period postoperatively, while healing occurs. This also allows patient controlled nocturnal deactivation, if desired. Nocturnal deactivation may decrease the incidence of cuff erosion in patients who have received pelvic irradiation.9 Deactivation is achieved by squeezing and releasing the pump as often as necessary to empty the cuff. Each squeeze empties the contents of the pump into the balloon, which has the effect of sucking an equivalent volume from the cuff into the pump until the cuff is empty. The pump empties last and then starts to refill (slowly because of the delay resistor) from the pressure regulating balloon. The deactivation button on the upper part of the pump is then pressed firmly as soon as the pump has refilled and, therefore, just before the cuff refills. This prevents the flow of fluid through the pump into the cuff, which remains empty. For activation the lower part of the pump is given a forceful squeeze, which moves the deactivation poppet back to the activated position. MICTURITION
Following activation the patient can deflate the cuff as desired to micturate. This is done by squeezing the pump until the cuff is empty and the pump no longer fills immediately. Delay resistors in the control mechanism in the upper part of the control pump allow the cuff to refill but only slowly for approximately 2 minutes, allowing sufficient time for the patient to void to completion before continence is restored. SURGICAL PRINCIPLES
All patients require broad-spectrum prophylactic antibiotics to avoid device infection, which would be a disaster. Our patients receive cefuroxime, metronidazole and gentamicin at anesthesia induction with the continuation of cefuroxime and metronidazole for 5 days, which combats aerobic and anaerobic pathogens. Aseptic surgical technique is essential. All patients receive Naseptin® cream nasally at 6-hour intervals and HibiscrubTM washes to the perineal area twice daily for a minimum of 24 hours preoperatively to minimize the risk of methicillin resistant Staphylococcus aureus infection. A rectal washout is done the night before and on the morning of surgery to ensure that the bowel is empty. The surgical technique has been described in detail by Scott et al10 and by one of us (ARM).11 The bladder neck is the position of choice for cuff placement in males, unless precluded by previous surgery, trauma or disease, and it is the only site available in women. The cuff may be inserted around the bulbar urethra in males with post-prostatectomy incontinence but the incidence of persistent stress incontinence is higher.12 Hence, we prefer the bladder neck. The cuff can be inserted around the bladder neck in females suprapubically or vaginally but higher infection rates have been reported with the vaginal approach.13 Thus, most surgeons favor the suprapubic extraperitoneal approach. Insertion of the AUS around the bladder neck in women is difficult. No natural plane exists posteriorly between the bladder and the vagina, and so a plane must be created between them, which carries a significant risk of perforating the vagina or urethra, leading to prosthesis infection. In addition, these patients have often undergone previous bladder neck procedures, making surgery more difficult. The pump is implanted into the labium majus of the female or in a dartos pouch in the scrotum of the male. The device is inserted in the deactivated state and it is activated after 2 to
6 weeks.14 Ideally a urethral catheter is avoided but if necessary a 12Fr silicone catheter may be inserted overnight or left for a few days in certain circumstances until normal voiding is resumed or clean intermittent self-catheterization (CISC) is commenced. The majority of patients remain incontinent until activation and, therefore, they need to continue whatever method they used to stay dry preoperatively. RESULTS
Recent analysis of all published results of Hajivassiliou relating to the AUS indicated that the overall continence rate is 73% with improved continence in 88% of cases.15 These results include approximately 10% of the devices implanted world wide. LONG-TERM RESULTS
Until recently no long-term data were published on the AUS. Levesque et al reported 10-year followup in 36 children and found a total of 22 functioning devices (61%) and a mean survival of 12.1 years.16 Only a third of cases had required no further surgery. In a further report of 32 children17 18 (56%) had a functioning AUS greater than 10 years after implantation and were satisfactorily continent. A total of 13 devices (41%) had been removed for infection/erosion and 19 (59%) had required revisions. There was a decrease in the revision rate of the AMS 800 after 1987 following a number of technical improvements, including a narrow back cuff design, a change in the synthetic material used to decrease fracturing and the introduction of the suture-free tubing connectors. A group at the Mayo Clinic reported a decrease in the reoperation rate from 42% to 17% following the introduction of these changes.18 Apart from our results there has been 1 other report of 10-year results in adults.19 More than half of these patients (56%) had an AUS inserted for neurogenic bladder dysfunction. Only 13% of the patients had the original AUS in situ but 61% were continent with revisions and 75% were continent when those who had died with a functioning AUS in situ were included. In our review of 100 patients with an AUS inserted more than 10 years ago the main indications were neurogenic bladder dysfunction in 58% and PPI in 24%. A third of the group (36%) had the original AUS in situ with satisfactory replacement in a further third at a mean of 8 years. Of the patients 21% had the AUS removed due to infection or erosion and subsequently had it successfully replaced or were dry without a device. However, overall in the 10-year period 37% of the devices were explanted due to infection or erosion, mainly in patients with complex stress incontinence who were incontinent despite the device. Although the numbers in this group are too small to draw firm conclusions, female patients with neurogenic bladder dysfunction also had a higher rate of removal than males (50% vs 30%), indicating that female patients are more prone to erosion because of difficulties in dissection, as described. Overall these results show a satisfactory long-term outcome for the AUS. At more than 10 years of followup our patients had an overall continence rate of 84%.7 PPI
Postoperative incontinence rates following radical perineal or retropubic prostatectomy are 0.5% to 40%.20 After transurethral resection of the prostate 1.7% of patients have stress incontinence and 0.4% are totally incontinent.21 The first report of the AUS for PPI was in 1983.22 In this early report of 66 patients using the AMS 791, 75% were continent. The rate was 93% when patients who had received radiotherapy were excluded. In an attempt to improve the continence rate, Britto et al suggested the addition of a sec-
CURRENT ROLE OF ARTIFICIAL URINARY SPHINCTER FOR INCONTINENCE
ond cuff with success rates greater than 95% in the short term.23 This double cuff technique was also reported to be successful by others.24 With more followup now available, the results of AUS insertion following PPI can be analyzed in the mid and long term. Singh and Thomas reported a social continence rate, defined as requiring 1 protective pad or less daily, of 96% in 28 patients at 3 years.25 The Mayo Clinic group reported a midterm continence rate of 88% in 323 patients with a mean followup of 6.5 years, of whom 70% had an AUS inserted for PPI.18 Revisions were reported in a third of the patients. Despite this significant revision rate patient satisfaction with the AUS for PPI is high, reported to be as high as 90%26 and increasing to 96% in those with post-transurethral prostate resection incontinence.27 Although postoperative continence is not 100%, the relative improvement in continence is the most significant factor affecting patient perceived outcome.28 Our 10-year results indicate continence in 91% of cses.7 Furthermore, the risk of removing the device because of infection or erosion was only 17% in the PPI group. However, the risk of early revision is higher if the cuff is placed around the bulbar urethra compared with the bladder neck (33% vs 16%). At our institution PPI is treated with a bulbar cuff rather than a bladder neck cuff because placement at this site is technically easier than at the latter.
NEUROGENIC BLADDER DYSFUNCTION
The AUS has revolutionized the quality of life of patients with neurogenic bladder dysfunction, most commonly those with spina bifida. It is important to ensure that these patients have an adequate capacity, stable bladder prior to sphincter insertion. Continence is most commonly achieved in this group by a combination of augmentation cystoplasty to control detrusor hyperreflexia and poor bladder compliance, and an AUS to control sphincter weakness. Detrusor hyperreflexia is a much more common problem than sphincter weakness and when the 2 conditions occur together, the latter is often difficult to identify. When it seems likely that the 2 conditions coexist, it is safe to perform augmentation cystoplasty and at the same time implant an AUS cuff in patients thought likely to need a sphincter to achieve full continence.29 It is then a relatively minor procedure to implant the balloon and pump if required. Approximately a third of the patient with spina bifida and an AUS require bladder augmentation in the long term.30 There is no significantly increased risk of AUS infection with simultaneous cystoplasty.3 All patients must be able to perform CISC because poor bladder emptying is usual and because they are at risk for bladder perforation if they fail to do so.31 CISC is required in 85% of those with neuropathy and augmentation cystoplasty alone, and in 100% of those with an AUS and augmentation cystoplasty.7 Stephenson and Mundy reported results in 61 patients with neurogenic bladder dysfunction using a combination of enterocystoplasty and an AUS if needed.29 Continence was achieved in 58 patients (95%). Midterm results were reported by Singh and Thomas with success in 92% of cases and reoperation in 28%.32 The 10-year continence rate in our patients with neurogenic bladder dysfunction was 86%.7 One of the main concerns about AUS use for neurogenic bladder dysfunction is an increase in detrusor hyperreflexia and a decrease in bladder compliance, which is associated with the increase in outlet resistance produced by an AUS without concomitant bladder augmentation. This may lead to upper tract deterioration. However, with careful postoperative surveillance by 6-month outpatient review the risk of renal damage can be minimized. Scott reviewed the records of 120 patients with neurogenic bladder dysfunction and a
421
mean followup of 36.8 months, and noted that in 90% the upper urinary tracts remained unchanged.33 CHILDREN
The AUS has been used extensively in children, mainly in those with neurogenic bladder dysfunction. In 1984 Light and Scott reported 90% success with the AS-792 in 132 children during 5 years, of whom 24% required revision surgery.34 Similar results have been reported with the AMS 800.35 Other indications are the exstrophy-epispadias complex,36 cloacal anomalies and post-trauma incontinence. The AUS has also allowed patients with poorly functioning or nonfunctioning urethral sphincters to be considered for undiversion when previously it was not possible to achieve urethral continence. STRESS INCONTINENCE IN FEMALES
The use of the artificial sphincter for primary treatment in women with severe, recurrent type III stress incontinence (ISD) is advocated by some groups.37 Others contest this and believe that conventional methods should be the treatment of choice.38 Results have been variable with up to 92% continence achieved when it is used as the primary method for treating ISD at 2.5 years39 but as low as 60% in our hands when used for treating the multiply operated patient.7 However, it should be noted that there are relatively few publications of the use of AUS for stress incontinence with generally small numbers, making it difficult to fully assess the real results of the AUS in this group. In our recent review of 68 female patients who had an AUS placed for ISD 37% retained the original sphincter at a median followup of 7 years and 16% had undergone sphincter replacement for late mechanical failure at a median of 5.5 years.40 Of the 63% of women in whom the sphincter was removed infection or erosion was the cause in 47%. Of this group 35% were dry without further surgery, while 39% underwent urinary diversion. The remaining 26% underwent successful sphincter replacement. COMPLICATIONS
AUS complications can be divided into mechanical complications, erosion/infection and recurrent incontinence. Mechanical. Mechanical complications are almost all surgery related with physical failure of the device occurring in less than 3%. The majority of mechanical failures (33%) are due to inadequate balloon pressure. Blocking or kinking of the tubing causes failure in 16% of cases.41 In patients who have the former problem and a bulbar cuff of less than 5.5 cm balloon pressure upgrade to the next highest pressure category may be enough. For cuffs of greater than 5.5 cm size; increasing the balloon pressure and reexploration of the cuff with down grading of cuff size by 0.5 cm is recommended.42 The mechanical failure rate reported in all publications is 12%15 with a decrease since the introduction of a narrow backing cuff to 7.6%.43 The need to increase balloon pressure was the most common cause of revision in our earlier patients. This rate has decreased with experience. Infection and erosion. Infection of the prosthesis is a disaster because the device must be removed. However, reinsertion may be possible at a later date. Infection may be introduced at implantation by atmospheric pathogens, unrecognized urinary tract infection, skin pathogens, or perforation of the urethra or vagina during surgery. Cuff erosion results from the pressure of the prosthesis on the underlying tissues. Any pathological condition, such as irradiation, previous surgery or trauma, or urethral scarring, which significantly decreases the vascularity of this tissue increases the risk of erosion. Cystoscopy enables the assessment of mucosal vascularity but for more accurate assessment Doppler flow studies may be used. Erosion may occur immediately or
422
CURRENT ROLE OF ARTIFICIAL URINARY SPHINCTER FOR INCONTINENCE
at any time after insertion (late erosion).44 It is not always possible to be certain whether erosion is the result of early infection or of pressure necrosis but the end result is the same. The device must be removed. Following explantation an 18Fr or 20Fr silicone catheter is placed for 4 to 6 weeks until there is radiological evidence of complete healing. The stricture rate after cuff erosion is remarkably low.15 The reported risk of early infection/erosion is about 4.5%15 but the risk is higher in neuropathic cases (15%), after failed sling surgery (67%) and after radical pelvic radiotherapy (57%).11 The incidence of erosion and infection can be decreased by ensuring that urine is sterile, antibiotic prophylaxis is given and surgical technique is meticulous. Late erosion. Late erosion was first described in 1993 by Duncan et al.45 It is an additional problem, particularly in patients with long life expectancy, such as children or young adults. Our results indicate a late erosion rate of 15%, occurring maximally 7 years after implantation, presumably due to urethral atrophy caused by cuff pressure.7 Recurrent incontinence. Incontinence immediately after device insertion is usually due to failure to use the device properly and specifically due to failure to deflate the cuff fully with incomplete bladder emptying and overflow incontinence. Delayed recurrent incontinence may be attributable to urethral atrophy beneath the cuff. One of us (ARM) has previously reported that 7% of patients require reoperation for this problem.11 This may be remedied by using a smaller cuff. However, if the original cuff used was the smallest available (4 cm), an alternative approach must be sought. Implantation of a second urethral cuff may be successful. In a study of 5 men with recurrent stress incontinence despite AUS placement after radical retropubic prostatectomy satisfactory continence was achieved in 4 using the double cuff technique.24 Recently another technique was described to treat such cases.46 The cuff was placed transcorporeally to add bulk to the urethra, allowing larger cuff sizes to be used. The buttress of the tunica albuginea and corporeal body also protects the urethra superiorly, which is a common site of urethral thinning and erosion. The disadvantage of this technique is the risk of postoperative erectile dysfunction. It is generally reserved for those with preexisting erectile dysfunction. Urethral fibrosis may also cause delayed incontinence and may require replacement of the device using a higher pressure balloon. The development of detrusor instability as part of the bladder response to obstruction may also lead to incontinence. Thus, all patients should be investigated with urodynamics prior to replacing the AUS since pharmacological therapy may resolve the problem. In all cases of recurrent incontinence systematic investigation is recommended. Inflate-deflate x-rays confirm correct hydraulic functioning of the AUS.47 If contrast medium cannot be identified within the system, a leak has occurred, most commonly at the cuff. The exact location can only be ascertained during surgical exploration but leakage is almost always from the cuff. This is because with long-term use the cuff develops creases on the inside from repeat filling and emptying. This leads to weakening of the material and eventual perforation. If inflate-deflate x-rays are normal, urethroscopy excludes cuff erosion as a cause of recurrent incontinence. CONCLUSIONS
The AUS has now been in use for more than 30 years. It passed through a number of changes, culminating in the AMS 800, which was introduced in 1982 and underwent design changes in 1987. The use of the AUS for PPI is established with a social continence rate of around 90% in the short and long term. Complications, particularly erosion and infection, are much rarer in this group due to the bulbar
position of the sphincter cuff in the majority. No other treatment for this distressing problem is as successful. For neurogenic bladder dysfunction the AUS has again achieved high continence rates in the short and long term. The complication and revision rates in this group are much higher but, when no other effective treatment exists, patients are generally prepared to accept the need for revision to achieve continence. The use of the AUS for stress incontinence is controversial. It is probably not indicated except for ISD because the risk of revision is unacceptably high and alternative successful treatments exist for most patients with stress incontinence who have hypermobility rather than ISD. Almost all patients with stress incontinence considered candidates for an AUS have undergone previous bladder neck surgery and, therefore, they are at increased risk for erosion. Furthermore, female patients with neurogenic bladder dysfunction who have not undergone previous surgery appear to be at higher risk for erosion than males, raising the question of whether the design of the current AUS is optimal for the female patient. Early complications following insertion are usually due to technical failure. Late complications are usually due to a leak in the system secondary to cuff perforation. The technical failure rate decreases as surgical experience increases. The risk of infection and erosion remains and it can only be decreased by ensuring that all precautions are taken. OVERCOMING PROBLEMS IN THE FUTURE: A NEW SPHINCTER
Although the AMS 800 remains a reliable means of achieving urinary continence in well selected and fully counseled patients, its deficiencies should not be ignored. The AMS 800 is costly. Its price varies among countries. In the United Kingdom each AUS costs in the region of £4,000 but is more expensive in the rest of Europe and much cheaper in the United States. The price also depends on institution use with better deals available at experienced and high use centers. Inherent stress incontinence, however minimal, remains a depressing problem in some patients following implantation of the AMS 800. The transmission of fluid from balloon to cuff is designed to be slow to allow time to void while the cuff refills. Thus, when abdominal pressure increases suddenly, as during coughing, the AMS 800 is unable to respond by rapidly increasing cuff pressure.5 This results in stress incontinence. The cuff pressure of the AMS 800 can only be adjusted by reoperation to exchange pressure balloons. Furthermore, the pressure provided by the AMS 800 is poorly defined, lying within a range of 10 cm H2O. To change the cuff pressure the balloon pressure may have to be increased by 1 to 19 cm H2O. In addition, the AMS 800 cuff is manufactured on a flat template. This means that when it is formed into the cylindrical cuff it may crack, precipitating future cuff perforation. This happens due to 2 properties of the inner silicone leaflet, namely memory and creep. After it is placed in situ the silicone develops a number of folds as it conforms to the underlying urethra. As the cuff is inflated, the folds persist because of memory. When deflated, the folds contact with the outer monofilament polypropylene backing of the cuff. Friction develops between the 2 layers. This provokes creep to occur. The silicone molecules migrate away from each other, culminating in overall thinning of the cuff with eventual perforation.48 Another problem with the AMS 800 is that it incorporates several color coded components and multiple connections as well as various sizes and pressures of cuffs and balloons, respectively. Hence, a working knowledge of the device and the ability to assemble it intraoperatively is essential.
CURRENT ROLE OF ARTIFICIAL URINARY SPHINCTER FOR INCONTINENCE
FIG. 4. AUS pressure-volume relationship. AMS 800 functions in plateau range of curve, maintaining fairly constant but relatively high pressure. New device works during initial phase of curve and, hence, it is less likely to cause urethral erosion because lower pressures are exerted.
Finally, some women find that the labial position of the pump is a problem, especially since their hands may come into contact with urine during pump manipulation. To overcome many of the problems intrinsic to the AMS 800 we developed a patented sphincter device at our institution. This device is similar in appearance and function but the cuff is molded on a curved instead of on a flat template, producing a single curved compressive cushion, such that cracks are unlikely and it is less likely to perforate. Also, it has 2 pressure regulating balloons. The second balloon allows a rapid response to changes in intra-abdominal pressure (fig. 3). When intra-abdominal pressure increases, pressure is transmitted from 1 balloon to the next, producing a rapid transient increase in cuff pressure, thus, maintaining continence. This may be especially useful in patients who use crutches to walk because their intra-abdominal pressure often far exceeds balloon pressure. The pressure in the system is determined by the increasing slope of the pressure/volume curve rather than by the plateau. Hence, the volumes used are smaller and they can be adjusted by injection or withdrawal of the filling solution fluid using a hypodermic needle, which penetrates the skin into a self-sealing filling port.49 This does not require anesthesia and it confers a major advantage over the AMS 800 device, of which the pressure can only be adjusted by reoperation to exchange pressure balloons. The AMS 800 device functions in the plateau range of the curve, maintaining fairly constant but relatively high pressure. Our new device works during the initial phase of the curve. Hence, it is less likely to cause urethral erosion since lower pressures are exerted. A pilot study of 7 patients has shown that this new sphincter is able to maintain steady cuff pressure after it is optimized (fig. 4).50 Midterm and long-term results are awaited but early results are encouraging. M. Schenk, American Medical Systems, Inc., provided figure 2. Dr. S. Knight, Institute of Urology, London, United Kingdom provided figure 3. REFERENCES
1. Foley, F. E. B.: An A U S: a new device and operations for the control of enuresis and urinary incontinence. General considerations. J Urol, 58: 250, 1947 2. Scott, F. B., Bradley, W. E. and Timm, G. W.: Treatment of urinary incontinence by an implantable prosthetic sphincter. J Urol, 112: 75, 1974 3. Mundy, A. R. and Stephenson, T. P.: Selection of patients for implantation of the Brantley Scott artificial urinary sphincter.
423
Br J Urol, 56: 717, 1984 4. Petrou, S. P. and Barrett, D. M.: The expanded role for the artificial sphincter. AUA Update Series, vol. X, lesson 16, p. 122, 1991 5. Nurse, D. E. and Mundy, A. R.: One hundred artificial sphincters. Br J Urol, 61: 318, 1988 6. Greenwell, T. J., Venn, S. N. and Mundy, A. R.: Augmentation cystoplasty. BJU Int, 88: 511, 2001 7. Venn, S. N., Greenwell, T. J. and Mundy, A. R.: The long-term outcome of artificial urinary sphincters. J Urol, 164: 702, 2000 8. Scott, F. B.: The artificial urinary sphincter experience in adults. Urol Clin North Am, 16: 105, 1989 9. Marks, J. L. and Light, J. K.: Management of urinary incontinence after prostatectomy with the artificial urinary sphincter. J Urol, 142: 302, 1989 10. Scott, F. B., Bradley, W. E., Timm, G. W. and Kothari, D.: Treatment of incontinence secondary to myelodysplasia by an implantable prosthetic urinary sphincter. South Med J, 66: 987, 1973 11. Mundy, A. R.: Artificial sphincters. Br J Urol, 67: 225, 1991 12. Sidi, A. A., Sinha, B. and Gonzalez, R.: Treatment of urinary incontinence with an artificial sphincter: further experience with the AS791/792 device. J Urol, 131: 891, 1984 13. Barrett, D. M. and Goldwasser, B.: The AUS: current management philosophy. AUA Update Series, vol. 5, lesson 32, 1988 14. Lindner, A., Kaufman, J. J. and Raz, S.: Further experience with the artificial urinary sphincter. J Urol, 129: 962, 1983 15. Hajivassiliou, C. A.: A review of the complications and results of implantation of the AMS artificial urinary sphincter. Eur Urol, 35: 36, 1999 16. Levesque, P. E., Bauer, S. B., Atal, A., Zurakowski, D., Colodny, A., Peters, C. et al: Ten-year experience with the artificial urinary sphincter in children. J Urol, 156: 625, 1996 17. Kryger, J. V., Spencer Barthold, J., Fleming, P. and Gonzalez, R.: The outcome of artificial urinary sphincter placement after a mean 15-year follow-up in a paediatric population. BJU Int, 83: 1026, 1999 18. Elliott, D. S. and Barrett, D. M.: Mayo Clinic long-term analysis of the functional durability of the AMS 800 artificial urinary sphincter: a review of 323 cases. J Urol, 159: 1206, 1998 19. Fulford, S. C., Sutton, C., Bales, G., Hickling, M. and Stephenson, T. P.: The fate of the ‘modern’ artificial urinary sphincter with a follow-up of more than 10 years. Br J Urol, 79: 713, 1997 20. Schreiter, F.: Bulbar artificial sphincter. Eur Urol, 11: 294, 1985 21. Mebust, W. K.: A review of transurethral resection of the prostate complications and the AUA national cooperation study. AUA Update Series, vol. VIII, lesson 34, 1989 22. Barrett, D. M. and Furlow, W. L.: Radical prostatectomy incontinence and the AS791 artificial urinary sphincter. J Urol, 129: 528, 1983 23. Brito, C. G., Mulcahy, J. J., Mitchell, M. E. and Adams, M. C.: Use of a double cuff AMS800 urinary sphincter for severe stress incontinence. J Urol, 149: 283, 1993 24. Kabalin, J. N.: Addition of a second urethral cuff to enhance performance of the artificial urinary sphincter. J Urol, 156: 1302, 1996 25. Singh, G. and Thomas, D. G.: Artificial urinary sphincter for post-prostatectomy incontinence. Br J Urol, 77: 248, 1996 26. Gundian, J. C., Barrett, D. M. and Parulkar, B. G.: Mayo clinic experience with use of the AMS800 artificial urinary sphincter for urinary incontinence following radical prostatectomy. J Urol, 142: 1459, 1989 27. Gundian, J. C., Barrett, D. M. and Parulkar, B. G.: Mayo Clinic experience with the AS800 artificial urinary sphincter for urinary incontinence after transurethral resection of prostate or open prostatectomy. Urology, 41: 318, 1993 28. Litwiller, S. E., Kim, K. B., Fone, P. D., deVere White, R. W. and Stone, A. R.: Post-prostatectomy incontinence and the artificial urinary sphincter: a long-term study of patient satisfaction and criteria for success. J Urol, 156: 1975, 1996 29. Stephenson, T. P. and Mundy, A. R.: Treatment of the neuropathic bladder by enterocystoplasty and selective sphincterotomy or sphincter ablation and replacement. Br J Urol, 57: 27, 1985 30. Sidi, A. A., Reinberg, Y. and Gonzalez, R.: Comparison of artificial urinary sphincter implantation and bladder neck reconstruction in patients with neurogenic urinary incontinence.
424
CURRENT ROLE OF ARTIFICIAL URINARY SPHINCTER FOR INCONTINENCE
J Urol, 138: 1120, 1987 31. Sheiner, J. R. and Kaplan, G. W.: Spontaneous bladder rupture following enterocystoplasty. J Urol, 140: 1157, 1988 32. Singh, G. and Thomas, D. G.: Artificial urinary sphincter in patients with neurogenic bladder dysfunction. Br J Urol, 77: 252, 1996 33. Scott, F. B., Fishman, I. J. and Shabsigh, R.: The impact of the artificial urinary sphincter in the neurogenic bladder on the upper urinary tracts. J Urol, 136: 636, 1986 34. Light, J. K. and Scott, F. B.: The artificial urinary sphincter in children. Br J Urol, 56: 54, 1984 35. Barrett, D. M. and Parulkar, B. G.: The artificial sphincter (AS-800). Experience in children and young adults. Urol Clin North Am, 16: 119, 1989 36. Light, J. K. and Scott, F. B.: Treatment of the epispadiasexstrophy complex with AS792 artificial urinary sphincter. J Urol, 129: 738, 1983 37. Scott, F. B.: The artificial urinary sphincter. Experience in adults. Urol Clin North Am, 16: 105, 1989 38. Barrett, D. M. and Furlow, W. L.: Artificial urinary sphincter in the management of female incontinence. In: Female Urology. Edited by S. Raz. Philadelphia. W. B. Saunders Co., pp. 284 – 292, 1983 39. Webster, G. D., Perez, L. M., Khoury, J. M. and Timmons, S. L.: Management of type III stress urinary incontinence using artificial urinary sphincter. Urology, 39: 499, 1992 40. Thomas, K., Venn, S. N. and Mundy, A. R.: Outcome of the artificial urinary sphincter in female patients. J Urol, 167: 1720, 2002 41. Goldwasser, B., Furlow, W. L. and Barrett, D. M.: The model AS 800 artificial urinary sphincter: Mayo Clinic experience.
J Urol, 137: 668, 1987 42. Barrett, D. M., Furlow, W. L. and Goldwasser, B.: Artificial urinary sphincters. Urol Arrival, 2: 117, 1988 43. Motley, R. C. and Barrett, D. M.: Artificial urinary sphincter cuff erosion. Experience with reimplantation in 38 patients. Urology, 35: 215, 1990 44. Scott, F. B., Bradley, W. E. and Timm, G. W.: Treatment of urinary incontinence by an implantable prosthetic urinary sphincter. J Urol, 167: 1125, 2002 45. Duncan, H. J., McInerney, P. D. and Mundy, A. R.: Late erosion. A new complication of artificial urinary sphincters. Br J Urol, 72: 597, 1993 46. Guralnick, M. L., Miller, E., Toh, K. L. and Webster, G. D.: Transcorporal artificial urinary sphincter cuff placement in cases requiring revision for erosion and urethral atrophy. J Urol, 167: 2075, 2002 47. Furlow, W. L. and Barrett, D. M.: Recurrent or persistent urinary incontinence in patients with the artificial urinary sphincter: diagnostic considerations and management. J Urol, 133: 792, 1985 48. Light, J. K. and Reynolds, J. C.: Impact of the new cuff design on reliability of the AS800 artificial urinary sphincter. J Urol, 147: 609, 1992 49. Craggs, M. D., Chaffey, N. J. and Mundy, A. R.: A preliminary report on a new hydraulic sphincter for controlling urinary incontinence. J Med Eng Technol, 15: 58, 1991 50. Craggs, M. D., Knight, S., Mundy, A. R., Dunglison, N. and Susser, J.: Cuff pressure vs. continence: long-term clinical results of a prototype artificial urinary sphincter with conditional occlusion for the treatment of genuine stress incontinence. BJU Int, suppl., 90: 419, 2002
Vol. 174, 418 – 424, August 2005 Printed in U.S.A.
DOI: 10.1097/01.ju.0000165345.11199.98
Review Articles THE CURRENT ROLE OF THE ARTIFICIAL URINARY SPHINCTER FOR THE TREATMENT OF URINARY INCONTINENCE MAHREEN HUSSAIN,* TAMSIN J. GREENWELL, SUZIE N. VENN
AND
ANTHONY R. MUNDY
From the Institute of Urology and Nephrology, University College London, London, United Kingdom
ABSTRACT
Purpose: The introduction of the artificial urinary sphincter (AUS) in 1972 was heralded as a revolution for the treatment of genuine stress incontinence. Initial enthusiasm was tempered by disappointment as complications occurred. The device has now been in routine clinical use for more than 30 years, and the indications and surgical principles involved in its use along with short-term and long-term outcomes are more clearly defined. Hence, we reviewed the literature to clarify the role of the AUS and offer a possible solution to its problems in the guise of a new sphincter. Materials and Methods: A MEDLINE search was performed and all articles relating to the role of the AUS for the treatment of urinary incontinence were reviewed. Results: The AMS 800 (American Medical Systems, Minnetonka, Minnesota) provides urinary continence in 73% of cases (range 61% to 96%) and it has a complication rate of 12% (range 3% to 33%) for mechanical failure, 4.5% to 67% for early infection/erosion, 15% for late erosion and 7% for delayed recurrent incontinence. The literature supports the role of the AUS as an important and reliable treatment modality for stress urinary incontinence and intrinsic sphincter deficiency. However, it is not suitable in all patients and its use for the management of hypermobility is controversial. Hence, careful patient selection according to indication is required with full preoperative counseling. Conclusions: Despite its reliability for achieving urinary continence the AMS 800 is not perfect. Newer devices, such as that being developed at our institution, may offer improved outcomes and decreased complication rates. KEY WORDS: urethra; urinary incontinence, stress; urinary sphincter, artificial; incontinence; complications HISTORY
The first description of an artificial urinary sphincter (AUS) in 1947 is credited to Foley,1 who developed an inflatable cuff wrapped around the penile urethra in a tunnel lined by skin and controlled by a detachable pump carried in the pocket. Scott implanted the first AUS in current use. The initial model, the AMS 721 (American Medical Systems), was introduced in 1972 (fig. 1).2 Valves controlled the direction of flow in the system and the pressure within it. Developments in design resulted in several changes, of which the most significant was the use of a balloon instead of valves to regulate pressure. Later models also saw the introduction of an entirely silicone cuff instead of a Dacron® reinforced silicone rubber cuff. A decrease in the number of components and connections culminated in the current 3-part AMS 800, which incorporates high quality aerospace components as a consequence of a collaboration between AMS and the National Aeronautics and Space Administration (fig. 2). First introduced in 1982, this has been the only AUS on the market for the last 20 years and more than 20,000 have been implanted.
FIG. 1. Original AMS 721 prosthetic urinary sphincter was first one to be implanted. It consisted of 4 main parts, namely reservoir, inflatable cuff and 2 pumping mechanisms for inflation and deflation, respectively.2
Submitted for publication June 2, 2004. * Correspondence: Institute of Urology and Nephrology, Middlesex INDICATIONS Hospital, 48 Riding House St., London W1W7EY, United Kingdom Our main indications for AUS insertion are post(telephone: 0044 207 679 9112; FAX: 0044 207 679 9584; e-mail: [email protected]). prostatectomy incontinence (PPI), sphincter weakness incon418
CURRENT ROLE OF ARTIFICIAL URINARY SPHINCTER FOR INCONTINENCE
419
AUS insertion. They require urodynamic and renal function assessment before and after implantation. Similar risks occur in patients with decreased compliance and small functional bladder capacity.5 The 2 groups should be treated with anticholinergic medication and, if this fails, augmentation cystoplasty should proceed6 to decrease contractility and increase compliance and functional bladder capacity simultaneously or before AUS insertion. CONTRAINDICATIONS
An absolute contraindication to the AUS is patient inability to manipulate the pump. AUS implantation should be avoided in patients at high risk for complications, eg women who have received radiotherapy to the pelvis. In our series the failure rate in this group was 100%.7 The AUS can be implanted around the bulbar urethra in men following radiotherapy to the pelvis with reasonable results, although the complication rate is higher. Previous bladder neck trauma, such as urethral injury in women, also carries a high risk of erosion. THE DEVICE
FIG. 2. Three-part AMS 800
Indications for AUS insertion3, 4 Dysfunction Congenital neurogenic: Spina bifida Sacral agenesis Shy-Drager syndrome Acquired neurogenic (spinal cord injury, multiple sclerosis) Congenital nonneurogenic (exstrophy-epispadias complex, cloaca, urogenital sinus complex) Acquired nonneurogenic: PPI Failed female anti-incontinence surgery Pelvic fracture
% Pts ⫾ 42 ⫾ 40 11 5 42 ⫾ 25 ⫾ 10 ⫾8
The AUS consists of 3 basic components, namely a pressure regulating balloon, an inflatable cuff and a control pump. The entire system is fluid filled and it functions hydraulically. Pressure in the system and, therefore, the occlusive force of the cuff is determined by the pressure regulating balloon. The balloon is made of a silicone polymer manufactured by a dip coating process,8 rendering it resistant to wear and tear. The pressure produced depends on the elasticity in the balloon wall, which in turn depends on the thickness of the wall, and the amount of fluid within it. The pressurevolume relationship is linear until it attains a plateau at between 16 and 24 ml of fluid, where it remains predictable within a narrow range (fig. 3).4 Pressure in the system is maintained except when the pump is squeezed. The choice of pressure is made by the surgeon intraoperatively. The ranges are 51 to 60, 61 to 70 and 71 to 80 cm H2O, while 41 to 50 and 81 to 90 cm H2O pressure cuffs are also available on special request. Cuffs are available in various sizes. All are 2 cm wide but they range from 4.0 to 11 cm long, allowing correct sizing of the bulbar urethra or bladder neck. To aid assembly the tubing is color coded. Tubing connecting the cuff to the pump is white and that connecting the balloon to the pump is black.
tinence due to neurogenic bladder dysfunction, intrinsic sphincter deficiency (ISD) and rare congenital causes of incontinence (see table).3, 4 PATIENT SELECTION AND ASSESSMENT
Full patient evaluation is essential prior to AUS insertion. It is important to establish that incontinence is sufficiently disabling to warrant surgical intervention. Urine must be sent for microscopy, culture and sensitivity, and sterility must be ensured prior to implantation. If cutaneous manifestations of incontinence such as ammonia dermatitis are present, a catheter should be placed for 2 weeks preoperatively to ensure intact skin with normal bacteriological flora at the time of sphincter implantation.4 It should then be removed shortly before surgery to decrease the risk of bacterial contamination by this route. Patient pre-assessment should be tailored to the indication. In those with PPI cystoscopy and urodynamics exclude coexisting bladder outlet obstruction or detrusor overactivity. If present, they must be treated prior to implanting an AUS. Those with neurogenic bladder dysfunction who have low volume detrusor hyperreflexia may be at risk for upper tract damage due to higher detrusor contractile pressure following
FIG. 3. New sphincter
420
CURRENT ROLE OF ARTIFICIAL URINARY SPHINCTER FOR INCONTINENCE DEACTIVATION AND ACTIVATION
At the end of the implantation procedure the device is deactivated. The control pump incorporates a valve, a refill delay resistor and a deactivation button. When this button is pressed, it stops fluid from being transferred among the 3 components, allowing the surgeon to maintain the sphincter cuff in a deflated state for a period postoperatively, while healing occurs. This also allows patient controlled nocturnal deactivation, if desired. Nocturnal deactivation may decrease the incidence of cuff erosion in patients who have received pelvic irradiation.9 Deactivation is achieved by squeezing and releasing the pump as often as necessary to empty the cuff. Each squeeze empties the contents of the pump into the balloon, which has the effect of sucking an equivalent volume from the cuff into the pump until the cuff is empty. The pump empties last and then starts to refill (slowly because of the delay resistor) from the pressure regulating balloon. The deactivation button on the upper part of the pump is then pressed firmly as soon as the pump has refilled and, therefore, just before the cuff refills. This prevents the flow of fluid through the pump into the cuff, which remains empty. For activation the lower part of the pump is given a forceful squeeze, which moves the deactivation poppet back to the activated position. MICTURITION
Following activation the patient can deflate the cuff as desired to micturate. This is done by squeezing the pump until the cuff is empty and the pump no longer fills immediately. Delay resistors in the control mechanism in the upper part of the control pump allow the cuff to refill but only slowly for approximately 2 minutes, allowing sufficient time for the patient to void to completion before continence is restored. SURGICAL PRINCIPLES
All patients require broad-spectrum prophylactic antibiotics to avoid device infection, which would be a disaster. Our patients receive cefuroxime, metronidazole and gentamicin at anesthesia induction with the continuation of cefuroxime and metronidazole for 5 days, which combats aerobic and anaerobic pathogens. Aseptic surgical technique is essential. All patients receive Naseptin® cream nasally at 6-hour intervals and HibiscrubTM washes to the perineal area twice daily for a minimum of 24 hours preoperatively to minimize the risk of methicillin resistant Staphylococcus aureus infection. A rectal washout is done the night before and on the morning of surgery to ensure that the bowel is empty. The surgical technique has been described in detail by Scott et al10 and by one of us (ARM).11 The bladder neck is the position of choice for cuff placement in males, unless precluded by previous surgery, trauma or disease, and it is the only site available in women. The cuff may be inserted around the bulbar urethra in males with post-prostatectomy incontinence but the incidence of persistent stress incontinence is higher.12 Hence, we prefer the bladder neck. The cuff can be inserted around the bladder neck in females suprapubically or vaginally but higher infection rates have been reported with the vaginal approach.13 Thus, most surgeons favor the suprapubic extraperitoneal approach. Insertion of the AUS around the bladder neck in women is difficult. No natural plane exists posteriorly between the bladder and the vagina, and so a plane must be created between them, which carries a significant risk of perforating the vagina or urethra, leading to prosthesis infection. In addition, these patients have often undergone previous bladder neck procedures, making surgery more difficult. The pump is implanted into the labium majus of the female or in a dartos pouch in the scrotum of the male. The device is inserted in the deactivated state and it is activated after 2 to
6 weeks.14 Ideally a urethral catheter is avoided but if necessary a 12Fr silicone catheter may be inserted overnight or left for a few days in certain circumstances until normal voiding is resumed or clean intermittent self-catheterization (CISC) is commenced. The majority of patients remain incontinent until activation and, therefore, they need to continue whatever method they used to stay dry preoperatively. RESULTS
Recent analysis of all published results of Hajivassiliou relating to the AUS indicated that the overall continence rate is 73% with improved continence in 88% of cases.15 These results include approximately 10% of the devices implanted world wide. LONG-TERM RESULTS
Until recently no long-term data were published on the AUS. Levesque et al reported 10-year followup in 36 children and found a total of 22 functioning devices (61%) and a mean survival of 12.1 years.16 Only a third of cases had required no further surgery. In a further report of 32 children17 18 (56%) had a functioning AUS greater than 10 years after implantation and were satisfactorily continent. A total of 13 devices (41%) had been removed for infection/erosion and 19 (59%) had required revisions. There was a decrease in the revision rate of the AMS 800 after 1987 following a number of technical improvements, including a narrow back cuff design, a change in the synthetic material used to decrease fracturing and the introduction of the suture-free tubing connectors. A group at the Mayo Clinic reported a decrease in the reoperation rate from 42% to 17% following the introduction of these changes.18 Apart from our results there has been 1 other report of 10-year results in adults.19 More than half of these patients (56%) had an AUS inserted for neurogenic bladder dysfunction. Only 13% of the patients had the original AUS in situ but 61% were continent with revisions and 75% were continent when those who had died with a functioning AUS in situ were included. In our review of 100 patients with an AUS inserted more than 10 years ago the main indications were neurogenic bladder dysfunction in 58% and PPI in 24%. A third of the group (36%) had the original AUS in situ with satisfactory replacement in a further third at a mean of 8 years. Of the patients 21% had the AUS removed due to infection or erosion and subsequently had it successfully replaced or were dry without a device. However, overall in the 10-year period 37% of the devices were explanted due to infection or erosion, mainly in patients with complex stress incontinence who were incontinent despite the device. Although the numbers in this group are too small to draw firm conclusions, female patients with neurogenic bladder dysfunction also had a higher rate of removal than males (50% vs 30%), indicating that female patients are more prone to erosion because of difficulties in dissection, as described. Overall these results show a satisfactory long-term outcome for the AUS. At more than 10 years of followup our patients had an overall continence rate of 84%.7 PPI
Postoperative incontinence rates following radical perineal or retropubic prostatectomy are 0.5% to 40%.20 After transurethral resection of the prostate 1.7% of patients have stress incontinence and 0.4% are totally incontinent.21 The first report of the AUS for PPI was in 1983.22 In this early report of 66 patients using the AMS 791, 75% were continent. The rate was 93% when patients who had received radiotherapy were excluded. In an attempt to improve the continence rate, Britto et al suggested the addition of a sec-
CURRENT ROLE OF ARTIFICIAL URINARY SPHINCTER FOR INCONTINENCE
ond cuff with success rates greater than 95% in the short term.23 This double cuff technique was also reported to be successful by others.24 With more followup now available, the results of AUS insertion following PPI can be analyzed in the mid and long term. Singh and Thomas reported a social continence rate, defined as requiring 1 protective pad or less daily, of 96% in 28 patients at 3 years.25 The Mayo Clinic group reported a midterm continence rate of 88% in 323 patients with a mean followup of 6.5 years, of whom 70% had an AUS inserted for PPI.18 Revisions were reported in a third of the patients. Despite this significant revision rate patient satisfaction with the AUS for PPI is high, reported to be as high as 90%26 and increasing to 96% in those with post-transurethral prostate resection incontinence.27 Although postoperative continence is not 100%, the relative improvement in continence is the most significant factor affecting patient perceived outcome.28 Our 10-year results indicate continence in 91% of cses.7 Furthermore, the risk of removing the device because of infection or erosion was only 17% in the PPI group. However, the risk of early revision is higher if the cuff is placed around the bulbar urethra compared with the bladder neck (33% vs 16%). At our institution PPI is treated with a bulbar cuff rather than a bladder neck cuff because placement at this site is technically easier than at the latter.
NEUROGENIC BLADDER DYSFUNCTION
The AUS has revolutionized the quality of life of patients with neurogenic bladder dysfunction, most commonly those with spina bifida. It is important to ensure that these patients have an adequate capacity, stable bladder prior to sphincter insertion. Continence is most commonly achieved in this group by a combination of augmentation cystoplasty to control detrusor hyperreflexia and poor bladder compliance, and an AUS to control sphincter weakness. Detrusor hyperreflexia is a much more common problem than sphincter weakness and when the 2 conditions occur together, the latter is often difficult to identify. When it seems likely that the 2 conditions coexist, it is safe to perform augmentation cystoplasty and at the same time implant an AUS cuff in patients thought likely to need a sphincter to achieve full continence.29 It is then a relatively minor procedure to implant the balloon and pump if required. Approximately a third of the patient with spina bifida and an AUS require bladder augmentation in the long term.30 There is no significantly increased risk of AUS infection with simultaneous cystoplasty.3 All patients must be able to perform CISC because poor bladder emptying is usual and because they are at risk for bladder perforation if they fail to do so.31 CISC is required in 85% of those with neuropathy and augmentation cystoplasty alone, and in 100% of those with an AUS and augmentation cystoplasty.7 Stephenson and Mundy reported results in 61 patients with neurogenic bladder dysfunction using a combination of enterocystoplasty and an AUS if needed.29 Continence was achieved in 58 patients (95%). Midterm results were reported by Singh and Thomas with success in 92% of cases and reoperation in 28%.32 The 10-year continence rate in our patients with neurogenic bladder dysfunction was 86%.7 One of the main concerns about AUS use for neurogenic bladder dysfunction is an increase in detrusor hyperreflexia and a decrease in bladder compliance, which is associated with the increase in outlet resistance produced by an AUS without concomitant bladder augmentation. This may lead to upper tract deterioration. However, with careful postoperative surveillance by 6-month outpatient review the risk of renal damage can be minimized. Scott reviewed the records of 120 patients with neurogenic bladder dysfunction and a
421
mean followup of 36.8 months, and noted that in 90% the upper urinary tracts remained unchanged.33 CHILDREN
The AUS has been used extensively in children, mainly in those with neurogenic bladder dysfunction. In 1984 Light and Scott reported 90% success with the AS-792 in 132 children during 5 years, of whom 24% required revision surgery.34 Similar results have been reported with the AMS 800.35 Other indications are the exstrophy-epispadias complex,36 cloacal anomalies and post-trauma incontinence. The AUS has also allowed patients with poorly functioning or nonfunctioning urethral sphincters to be considered for undiversion when previously it was not possible to achieve urethral continence. STRESS INCONTINENCE IN FEMALES
The use of the artificial sphincter for primary treatment in women with severe, recurrent type III stress incontinence (ISD) is advocated by some groups.37 Others contest this and believe that conventional methods should be the treatment of choice.38 Results have been variable with up to 92% continence achieved when it is used as the primary method for treating ISD at 2.5 years39 but as low as 60% in our hands when used for treating the multiply operated patient.7 However, it should be noted that there are relatively few publications of the use of AUS for stress incontinence with generally small numbers, making it difficult to fully assess the real results of the AUS in this group. In our recent review of 68 female patients who had an AUS placed for ISD 37% retained the original sphincter at a median followup of 7 years and 16% had undergone sphincter replacement for late mechanical failure at a median of 5.5 years.40 Of the 63% of women in whom the sphincter was removed infection or erosion was the cause in 47%. Of this group 35% were dry without further surgery, while 39% underwent urinary diversion. The remaining 26% underwent successful sphincter replacement. COMPLICATIONS
AUS complications can be divided into mechanical complications, erosion/infection and recurrent incontinence. Mechanical. Mechanical complications are almost all surgery related with physical failure of the device occurring in less than 3%. The majority of mechanical failures (33%) are due to inadequate balloon pressure. Blocking or kinking of the tubing causes failure in 16% of cases.41 In patients who have the former problem and a bulbar cuff of less than 5.5 cm balloon pressure upgrade to the next highest pressure category may be enough. For cuffs of greater than 5.5 cm size; increasing the balloon pressure and reexploration of the cuff with down grading of cuff size by 0.5 cm is recommended.42 The mechanical failure rate reported in all publications is 12%15 with a decrease since the introduction of a narrow backing cuff to 7.6%.43 The need to increase balloon pressure was the most common cause of revision in our earlier patients. This rate has decreased with experience. Infection and erosion. Infection of the prosthesis is a disaster because the device must be removed. However, reinsertion may be possible at a later date. Infection may be introduced at implantation by atmospheric pathogens, unrecognized urinary tract infection, skin pathogens, or perforation of the urethra or vagina during surgery. Cuff erosion results from the pressure of the prosthesis on the underlying tissues. Any pathological condition, such as irradiation, previous surgery or trauma, or urethral scarring, which significantly decreases the vascularity of this tissue increases the risk of erosion. Cystoscopy enables the assessment of mucosal vascularity but for more accurate assessment Doppler flow studies may be used. Erosion may occur immediately or
422
CURRENT ROLE OF ARTIFICIAL URINARY SPHINCTER FOR INCONTINENCE
at any time after insertion (late erosion).44 It is not always possible to be certain whether erosion is the result of early infection or of pressure necrosis but the end result is the same. The device must be removed. Following explantation an 18Fr or 20Fr silicone catheter is placed for 4 to 6 weeks until there is radiological evidence of complete healing. The stricture rate after cuff erosion is remarkably low.15 The reported risk of early infection/erosion is about 4.5%15 but the risk is higher in neuropathic cases (15%), after failed sling surgery (67%) and after radical pelvic radiotherapy (57%).11 The incidence of erosion and infection can be decreased by ensuring that urine is sterile, antibiotic prophylaxis is given and surgical technique is meticulous. Late erosion. Late erosion was first described in 1993 by Duncan et al.45 It is an additional problem, particularly in patients with long life expectancy, such as children or young adults. Our results indicate a late erosion rate of 15%, occurring maximally 7 years after implantation, presumably due to urethral atrophy caused by cuff pressure.7 Recurrent incontinence. Incontinence immediately after device insertion is usually due to failure to use the device properly and specifically due to failure to deflate the cuff fully with incomplete bladder emptying and overflow incontinence. Delayed recurrent incontinence may be attributable to urethral atrophy beneath the cuff. One of us (ARM) has previously reported that 7% of patients require reoperation for this problem.11 This may be remedied by using a smaller cuff. However, if the original cuff used was the smallest available (4 cm), an alternative approach must be sought. Implantation of a second urethral cuff may be successful. In a study of 5 men with recurrent stress incontinence despite AUS placement after radical retropubic prostatectomy satisfactory continence was achieved in 4 using the double cuff technique.24 Recently another technique was described to treat such cases.46 The cuff was placed transcorporeally to add bulk to the urethra, allowing larger cuff sizes to be used. The buttress of the tunica albuginea and corporeal body also protects the urethra superiorly, which is a common site of urethral thinning and erosion. The disadvantage of this technique is the risk of postoperative erectile dysfunction. It is generally reserved for those with preexisting erectile dysfunction. Urethral fibrosis may also cause delayed incontinence and may require replacement of the device using a higher pressure balloon. The development of detrusor instability as part of the bladder response to obstruction may also lead to incontinence. Thus, all patients should be investigated with urodynamics prior to replacing the AUS since pharmacological therapy may resolve the problem. In all cases of recurrent incontinence systematic investigation is recommended. Inflate-deflate x-rays confirm correct hydraulic functioning of the AUS.47 If contrast medium cannot be identified within the system, a leak has occurred, most commonly at the cuff. The exact location can only be ascertained during surgical exploration but leakage is almost always from the cuff. This is because with long-term use the cuff develops creases on the inside from repeat filling and emptying. This leads to weakening of the material and eventual perforation. If inflate-deflate x-rays are normal, urethroscopy excludes cuff erosion as a cause of recurrent incontinence. CONCLUSIONS
The AUS has now been in use for more than 30 years. It passed through a number of changes, culminating in the AMS 800, which was introduced in 1982 and underwent design changes in 1987. The use of the AUS for PPI is established with a social continence rate of around 90% in the short and long term. Complications, particularly erosion and infection, are much rarer in this group due to the bulbar
position of the sphincter cuff in the majority. No other treatment for this distressing problem is as successful. For neurogenic bladder dysfunction the AUS has again achieved high continence rates in the short and long term. The complication and revision rates in this group are much higher but, when no other effective treatment exists, patients are generally prepared to accept the need for revision to achieve continence. The use of the AUS for stress incontinence is controversial. It is probably not indicated except for ISD because the risk of revision is unacceptably high and alternative successful treatments exist for most patients with stress incontinence who have hypermobility rather than ISD. Almost all patients with stress incontinence considered candidates for an AUS have undergone previous bladder neck surgery and, therefore, they are at increased risk for erosion. Furthermore, female patients with neurogenic bladder dysfunction who have not undergone previous surgery appear to be at higher risk for erosion than males, raising the question of whether the design of the current AUS is optimal for the female patient. Early complications following insertion are usually due to technical failure. Late complications are usually due to a leak in the system secondary to cuff perforation. The technical failure rate decreases as surgical experience increases. The risk of infection and erosion remains and it can only be decreased by ensuring that all precautions are taken. OVERCOMING PROBLEMS IN THE FUTURE: A NEW SPHINCTER
Although the AMS 800 remains a reliable means of achieving urinary continence in well selected and fully counseled patients, its deficiencies should not be ignored. The AMS 800 is costly. Its price varies among countries. In the United Kingdom each AUS costs in the region of £4,000 but is more expensive in the rest of Europe and much cheaper in the United States. The price also depends on institution use with better deals available at experienced and high use centers. Inherent stress incontinence, however minimal, remains a depressing problem in some patients following implantation of the AMS 800. The transmission of fluid from balloon to cuff is designed to be slow to allow time to void while the cuff refills. Thus, when abdominal pressure increases suddenly, as during coughing, the AMS 800 is unable to respond by rapidly increasing cuff pressure.5 This results in stress incontinence. The cuff pressure of the AMS 800 can only be adjusted by reoperation to exchange pressure balloons. Furthermore, the pressure provided by the AMS 800 is poorly defined, lying within a range of 10 cm H2O. To change the cuff pressure the balloon pressure may have to be increased by 1 to 19 cm H2O. In addition, the AMS 800 cuff is manufactured on a flat template. This means that when it is formed into the cylindrical cuff it may crack, precipitating future cuff perforation. This happens due to 2 properties of the inner silicone leaflet, namely memory and creep. After it is placed in situ the silicone develops a number of folds as it conforms to the underlying urethra. As the cuff is inflated, the folds persist because of memory. When deflated, the folds contact with the outer monofilament polypropylene backing of the cuff. Friction develops between the 2 layers. This provokes creep to occur. The silicone molecules migrate away from each other, culminating in overall thinning of the cuff with eventual perforation.48 Another problem with the AMS 800 is that it incorporates several color coded components and multiple connections as well as various sizes and pressures of cuffs and balloons, respectively. Hence, a working knowledge of the device and the ability to assemble it intraoperatively is essential.
CURRENT ROLE OF ARTIFICIAL URINARY SPHINCTER FOR INCONTINENCE
FIG. 4. AUS pressure-volume relationship. AMS 800 functions in plateau range of curve, maintaining fairly constant but relatively high pressure. New device works during initial phase of curve and, hence, it is less likely to cause urethral erosion because lower pressures are exerted.
Finally, some women find that the labial position of the pump is a problem, especially since their hands may come into contact with urine during pump manipulation. To overcome many of the problems intrinsic to the AMS 800 we developed a patented sphincter device at our institution. This device is similar in appearance and function but the cuff is molded on a curved instead of on a flat template, producing a single curved compressive cushion, such that cracks are unlikely and it is less likely to perforate. Also, it has 2 pressure regulating balloons. The second balloon allows a rapid response to changes in intra-abdominal pressure (fig. 3). When intra-abdominal pressure increases, pressure is transmitted from 1 balloon to the next, producing a rapid transient increase in cuff pressure, thus, maintaining continence. This may be especially useful in patients who use crutches to walk because their intra-abdominal pressure often far exceeds balloon pressure. The pressure in the system is determined by the increasing slope of the pressure/volume curve rather than by the plateau. Hence, the volumes used are smaller and they can be adjusted by injection or withdrawal of the filling solution fluid using a hypodermic needle, which penetrates the skin into a self-sealing filling port.49 This does not require anesthesia and it confers a major advantage over the AMS 800 device, of which the pressure can only be adjusted by reoperation to exchange pressure balloons. The AMS 800 device functions in the plateau range of the curve, maintaining fairly constant but relatively high pressure. Our new device works during the initial phase of the curve. Hence, it is less likely to cause urethral erosion since lower pressures are exerted. A pilot study of 7 patients has shown that this new sphincter is able to maintain steady cuff pressure after it is optimized (fig. 4).50 Midterm and long-term results are awaited but early results are encouraging. M. Schenk, American Medical Systems, Inc., provided figure 2. Dr. S. Knight, Institute of Urology, London, United Kingdom provided figure 3. REFERENCES
1. Foley, F. E. B.: An A U S: a new device and operations for the control of enuresis and urinary incontinence. General considerations. J Urol, 58: 250, 1947 2. Scott, F. B., Bradley, W. E. and Timm, G. W.: Treatment of urinary incontinence by an implantable prosthetic sphincter. J Urol, 112: 75, 1974 3. Mundy, A. R. and Stephenson, T. P.: Selection of patients for implantation of the Brantley Scott artificial urinary sphincter.
423
Br J Urol, 56: 717, 1984 4. Petrou, S. P. and Barrett, D. M.: The expanded role for the artificial sphincter. AUA Update Series, vol. X, lesson 16, p. 122, 1991 5. Nurse, D. E. and Mundy, A. R.: One hundred artificial sphincters. Br J Urol, 61: 318, 1988 6. Greenwell, T. J., Venn, S. N. and Mundy, A. R.: Augmentation cystoplasty. BJU Int, 88: 511, 2001 7. Venn, S. N., Greenwell, T. J. and Mundy, A. R.: The long-term outcome of artificial urinary sphincters. J Urol, 164: 702, 2000 8. Scott, F. B.: The artificial urinary sphincter experience in adults. Urol Clin North Am, 16: 105, 1989 9. Marks, J. L. and Light, J. K.: Management of urinary incontinence after prostatectomy with the artificial urinary sphincter. J Urol, 142: 302, 1989 10. Scott, F. B., Bradley, W. E., Timm, G. W. and Kothari, D.: Treatment of incontinence secondary to myelodysplasia by an implantable prosthetic urinary sphincter. South Med J, 66: 987, 1973 11. Mundy, A. R.: Artificial sphincters. Br J Urol, 67: 225, 1991 12. Sidi, A. A., Sinha, B. and Gonzalez, R.: Treatment of urinary incontinence with an artificial sphincter: further experience with the AS791/792 device. J Urol, 131: 891, 1984 13. Barrett, D. M. and Goldwasser, B.: The AUS: current management philosophy. AUA Update Series, vol. 5, lesson 32, 1988 14. Lindner, A., Kaufman, J. J. and Raz, S.: Further experience with the artificial urinary sphincter. J Urol, 129: 962, 1983 15. Hajivassiliou, C. A.: A review of the complications and results of implantation of the AMS artificial urinary sphincter. Eur Urol, 35: 36, 1999 16. Levesque, P. E., Bauer, S. B., Atal, A., Zurakowski, D., Colodny, A., Peters, C. et al: Ten-year experience with the artificial urinary sphincter in children. J Urol, 156: 625, 1996 17. Kryger, J. V., Spencer Barthold, J., Fleming, P. and Gonzalez, R.: The outcome of artificial urinary sphincter placement after a mean 15-year follow-up in a paediatric population. BJU Int, 83: 1026, 1999 18. Elliott, D. S. and Barrett, D. M.: Mayo Clinic long-term analysis of the functional durability of the AMS 800 artificial urinary sphincter: a review of 323 cases. J Urol, 159: 1206, 1998 19. Fulford, S. C., Sutton, C., Bales, G., Hickling, M. and Stephenson, T. P.: The fate of the ‘modern’ artificial urinary sphincter with a follow-up of more than 10 years. Br J Urol, 79: 713, 1997 20. Schreiter, F.: Bulbar artificial sphincter. Eur Urol, 11: 294, 1985 21. Mebust, W. K.: A review of transurethral resection of the prostate complications and the AUA national cooperation study. AUA Update Series, vol. VIII, lesson 34, 1989 22. Barrett, D. M. and Furlow, W. L.: Radical prostatectomy incontinence and the AS791 artificial urinary sphincter. J Urol, 129: 528, 1983 23. Brito, C. G., Mulcahy, J. J., Mitchell, M. E. and Adams, M. C.: Use of a double cuff AMS800 urinary sphincter for severe stress incontinence. J Urol, 149: 283, 1993 24. Kabalin, J. N.: Addition of a second urethral cuff to enhance performance of the artificial urinary sphincter. J Urol, 156: 1302, 1996 25. Singh, G. and Thomas, D. G.: Artificial urinary sphincter for post-prostatectomy incontinence. Br J Urol, 77: 248, 1996 26. Gundian, J. C., Barrett, D. M. and Parulkar, B. G.: Mayo clinic experience with use of the AMS800 artificial urinary sphincter for urinary incontinence following radical prostatectomy. J Urol, 142: 1459, 1989 27. Gundian, J. C., Barrett, D. M. and Parulkar, B. G.: Mayo Clinic experience with the AS800 artificial urinary sphincter for urinary incontinence after transurethral resection of prostate or open prostatectomy. Urology, 41: 318, 1993 28. Litwiller, S. E., Kim, K. B., Fone, P. D., deVere White, R. W. and Stone, A. R.: Post-prostatectomy incontinence and the artificial urinary sphincter: a long-term study of patient satisfaction and criteria for success. J Urol, 156: 1975, 1996 29. Stephenson, T. P. and Mundy, A. R.: Treatment of the neuropathic bladder by enterocystoplasty and selective sphincterotomy or sphincter ablation and replacement. Br J Urol, 57: 27, 1985 30. Sidi, A. A., Reinberg, Y. and Gonzalez, R.: Comparison of artificial urinary sphincter implantation and bladder neck reconstruction in patients with neurogenic urinary incontinence.
424
CURRENT ROLE OF ARTIFICIAL URINARY SPHINCTER FOR INCONTINENCE
J Urol, 138: 1120, 1987 31. Sheiner, J. R. and Kaplan, G. W.: Spontaneous bladder rupture following enterocystoplasty. J Urol, 140: 1157, 1988 32. Singh, G. and Thomas, D. G.: Artificial urinary sphincter in patients with neurogenic bladder dysfunction. Br J Urol, 77: 252, 1996 33. Scott, F. B., Fishman, I. J. and Shabsigh, R.: The impact of the artificial urinary sphincter in the neurogenic bladder on the upper urinary tracts. J Urol, 136: 636, 1986 34. Light, J. K. and Scott, F. B.: The artificial urinary sphincter in children. Br J Urol, 56: 54, 1984 35. Barrett, D. M. and Parulkar, B. G.: The artificial sphincter (AS-800). Experience in children and young adults. Urol Clin North Am, 16: 119, 1989 36. Light, J. K. and Scott, F. B.: Treatment of the epispadiasexstrophy complex with AS792 artificial urinary sphincter. J Urol, 129: 738, 1983 37. Scott, F. B.: The artificial urinary sphincter. Experience in adults. Urol Clin North Am, 16: 105, 1989 38. Barrett, D. M. and Furlow, W. L.: Artificial urinary sphincter in the management of female incontinence. In: Female Urology. Edited by S. Raz. Philadelphia. W. B. Saunders Co., pp. 284 – 292, 1983 39. Webster, G. D., Perez, L. M., Khoury, J. M. and Timmons, S. L.: Management of type III stress urinary incontinence using artificial urinary sphincter. Urology, 39: 499, 1992 40. Thomas, K., Venn, S. N. and Mundy, A. R.: Outcome of the artificial urinary sphincter in female patients. J Urol, 167: 1720, 2002 41. Goldwasser, B., Furlow, W. L. and Barrett, D. M.: The model AS 800 artificial urinary sphincter: Mayo Clinic experience.
J Urol, 137: 668, 1987 42. Barrett, D. M., Furlow, W. L. and Goldwasser, B.: Artificial urinary sphincters. Urol Arrival, 2: 117, 1988 43. Motley, R. C. and Barrett, D. M.: Artificial urinary sphincter cuff erosion. Experience with reimplantation in 38 patients. Urology, 35: 215, 1990 44. Scott, F. B., Bradley, W. E. and Timm, G. W.: Treatment of urinary incontinence by an implantable prosthetic urinary sphincter. J Urol, 167: 1125, 2002 45. Duncan, H. J., McInerney, P. D. and Mundy, A. R.: Late erosion. A new complication of artificial urinary sphincters. Br J Urol, 72: 597, 1993 46. Guralnick, M. L., Miller, E., Toh, K. L. and Webster, G. D.: Transcorporal artificial urinary sphincter cuff placement in cases requiring revision for erosion and urethral atrophy. J Urol, 167: 2075, 2002 47. Furlow, W. L. and Barrett, D. M.: Recurrent or persistent urinary incontinence in patients with the artificial urinary sphincter: diagnostic considerations and management. J Urol, 133: 792, 1985 48. Light, J. K. and Reynolds, J. C.: Impact of the new cuff design on reliability of the AS800 artificial urinary sphincter. J Urol, 147: 609, 1992 49. Craggs, M. D., Chaffey, N. J. and Mundy, A. R.: A preliminary report on a new hydraulic sphincter for controlling urinary incontinence. J Med Eng Technol, 15: 58, 1991 50. Craggs, M. D., Knight, S., Mundy, A. R., Dunglison, N. and Susser, J.: Cuff pressure vs. continence: long-term clinical results of a prototype artificial urinary sphincter with conditional occlusion for the treatment of genuine stress incontinence. BJU Int, suppl., 90: 419, 2002