Experimental evaluation of an electromechanical artificial urinary sphincter in an animal model


Correspondence: Massimo Valerio, Service d'Urologie, Centre Hospitalire Universitaire Vaudois, Rue du Bugnon 21, 1011 Lausanne-CHUV, Switzerland.

e-mail: massimo.valerio@chuv.ch


What's known on the subject? and What does the study add?

  • The AMS 800 urinary control system is the gold standard for the treatment of urinary incontinence due to sphincter insufficiency. Despite excellent functional outcome and latest technological improvements, the revision rate remains significant.
  • To overcome the shortcomings of the current device, we developed a modern electromechanical artificial urinary sphincter. The results demonstrated that this new sphincter is effective and well tolerated up to 3 months. This preliminary study represents a first step in the clinical application of novel technologies and an alternative compression mechanism to the urethra.


  • To evaluate the effectiveness in continence achievement of a new electromechanical artificial urinary sphincter (emAUS) in an animal model.
  • To assess urethral response and animal general response to short-term and mid-term activation of the emAUS.

Materials and Methods

  • The principle of the emAUS is electromechanical induction of alternating compression of successive segments of the urethra by a series of cuffs activated by artificial muscles. Between February 2009 and May 2010 the emAUS was implanted in 17 sheep divided into three groups.
  • The first phase aimed to measure bladder leak point pressure during the activation of the device.
  • The second and third phases aimed to assess tissue response to the presence of the device after 2–9 weeks and after 3 months respectively. Histopathological and immunohistochemistry evaluation of the urethra was performed.


  • Bladder leak point pressure was measured at levels between 1091 ± 30.6 cmH2O and 1244.1 ± 99 cmH2O (mean ± standard deviation) depending on the number of cuffs used.
  • At gross examination, the explanted urethra showed no sign of infection, atrophy or stricture.
  • On microscopic examination no significant difference in structure was found between urethral structure surrounded by a cuff and control urethra. In the peripheral tissues, the implanted material elicited a chronic foreign body reaction. Apart from one case, specimens did not show significant presence of lymphocytes, polymorphonuclear leucocytes, necrosis or cell degeneration.
  • Immunohistochemistry confirmed the absence of macrophages in the samples.


  • This animal study shows that the emAUS can provide continence.
  • This new electronic controlled sequential alternating compression mechanism can avoid damage to urethral vascularity, at least up to 3 months after implantation.
  • After this positive proof of concept, long-term studies are needed before clinical application could be considered.

American Medical Systems


artificial urinary sphincter


electromechanical AUS


bladder leak point pressure.


The AMS 800 (American Medical Systems, Minnetonka, MN, USA) artificial urinary sphincter (AUS) is currently the gold standard AUS device [1] used to treat stress urinary incontinence due to sphincter insufficiency when simpler standard techniques fail or are inappropriate. Results from a meta-analysis show that 73% of patients achieve full continence and 88% have improved continence [2]. However, up to 60% of the implanted devices require surgical revisions at 10 years due to technical failures inherent in the hydraulic mechanism or to ischaemic injury of the urethra [3, 4]. Thus, research into alternative solutions seems reasonable.

In this animal study we have tested the hypothesis that an electromechanical AUS (emAUS) might produce continence by applying synchronized sequential compression to the urethra. The purpose of this study was to test the effectiveness of the device, the tissue response after implantation and the animal tolerance.

Materials and Methods

Description of the emAUS System

The emAUS – in these experiments – consists of two parts: a contractile unit, including two or more urethral cuffs that provide compression, which is implanted intra-corporeally (Fig. 1), and an electronic board which controls it extra-corporeally. Materials used are polyoxymethylene for the cuffs and medical grade silicon to insulate the electric cables and the contractile unit. The closing mechanism of the device is a sequential compression of a segment of the urethra by two or more cuffs. Each cuff is independently activated by a specific contractile unit under the control of an integrated microprocessor. This process is called the piano concept (Fig. 2). The fibres of each contractile unit are composed of nitinol, a metal alloy of nickel and titanium. When the artificial muscles forming each contractile unit are relaxed, the urethra opens. Nitinol alloys exhibit two related properties: shape memory and pseudoelasticity. These refer to the deformation process of the fibres under specific conditions of heating. The nitinol fibre has a uniform crystal structure that radically changes at a specific temperature. When the fibre remains below this transition temperature it can be stretched and deformed without permanent damage (pseudoelasticity). If a permanent deformation of the alloy occurs at its transition temperature, the fibre needs to be overheated to reverse and recover its previous unstretched shape (shape memory). This process can be repeated indefinitely without damage. Some existing implantable devices using nitinol have already confirmed the biocompatibility of the system [5]. In our study we used a given current to heat the wire up to 70 °C which is its transitional temperature. The cooling process was passive due to heat dissipation into surrounding structures but not tissues.

Figure 1.

The emAUS consists of a series of two or three cuffs which are implanted connected to a contractile unit (shown here) and an electronic board (not shown) that controls it.

Figure 2.

Schematic cartoon of the piano concept. The sequential alternating compression of successive segments of the urethra should preserve urethral vascularity.

Study Design

An animal study was conducted between February 2009 and May 2010. The protocol was approved by our local ethics committee according to current Swiss regulations on animal experimentation. The emAUS was implanted in 17 adult orchidectomized male sheep. Sheep were used because they appear to be the most accurate animal model for AUS testing [6]. The study was undertaken in three different phases. The first phase aimed to show ‘proof of principle’ by showing that the device could produce sufficient bladder outflow obstruction to be a reasonable proxy for the achievement of continence. The second and third phases aimed to evaluate the urethral tissue response and the general tolerance of animals submitted to short-term and mid-term activation of the devices, respectively.

All procedures were performed by the same team. Animals received a parenteral antibiotic at induction of anaesthesia as a prophylactic measure. They were anaesthetized intravenously by a thiopentone–pentobarbitone mixture and then by isoflurane inhalation during the procedure. The urethra was exposed through a low midline incision; a plane was created around it, and the cuffs of the emAUS were positioned around the urethra over a length of 20–35 mm just distal to the external urethral sphincter (Fig. 3). The contractile unit was placed through a second incision (Fig. 4). The pulling force to close each cuff was 0.7 N in every case. This was considered to be the lowest force able to produce complete closure of the urethra based on previous ex vivo experimentation in this sheep model. In phases 2 and 3, the contractile unit was then connected to the external electronic board by cables passing through the subcutaneous tissue to the back of each animal.

Figure 3.

A series of two or three cuffs are positioned around the urethra.

Figure 4.

The contractile unit is positioned 5–10 cm away from the cuffs through a second lateral incision.

Phase 1

Between February and June 2009, a three-cuff emAUS was implanted in three sheep. The effectiveness of the device was tested by measuring the bladder leak point pressure (BLPP), defined as the lowest intravesical pressure at which urine leakage greater than 1 mL/min occurred as intravesical pressure was increased by continuous bladder infusion of saline solution up to maximum bladder capacity. After the measurement of BLPP with one cuff closed, measurements of BLPP with two and three cuffs closed synchronously were undertaken. Animals were then sacrificed.

Phase 2

Between July and September 2009 the same device used in phase 1 was positioned, as described above, in six sheep. The device was activated for 20 h/day. It was turned off for 10 min every hour to allow the animals to urinate. During the phase of activation, the cuffs were activated in a sequential mode, with each cuff squeezing its segment of the urethra for 10 min at a time. The system was regularly checked every week by a computer connected to the external cable. Any failure of any component of the emAUS – the connections, the cuffs, the nitinol fibres or the microprocessor – was recorded. At each period of 2, 5 and 9 weeks after implantation, two sheep were sacrificed in order to assess the tissue response. The urethra was macroscopically examined and cross-sections of the urethra within and adjacent to the cuffs were taken, embedded in paraffin and prepared for either haematoxylin and eosin staining or toluidine blue staining. Urethra from a remote site was used as a control.

Phase 3

Between January and February 2010 a two-cuff emAUS was implanted into eight sheep to compare the results with the three-cuff device used in phase 2. The study protocol was the same as in the second phase except that, in addition, the animals' pain or discomfort, urine volumes and vital parameters were assessed six times a day by appropriately trained staff. As the phase 2 study showed no adverse tissue response to the presence of the emAUS on histology it was felt that the use of a sham group as a basis for comparison was unnecessary.

After 12 weeks, sheep were weighed and killed by an intravenous injection of pentobarbital. Weight variation between the beginning and the end of the study was considered an index of health status. As in phase 2 the segment of the urethra related to the implanted cuffs and the tissues surrounding the control unit were excised and examined macroscopically. Ten serial transverse sections of each site were then prepared. The first section was stained with safranin–haematoxylin–eosin, the second section was stained with Masson's trichrome and the remaining eight sections were used for macrophage immunolabelling. Mesenteric lymph nodes were used as positive control. The presence of polymorphonuclear leucocytes, lymphocytes, necrosis, encapsulation, cell degeneration and immunoreactivity of macrophages were separately evaluated in each specimen by a single pathologist. Values were expressed as a score from 0 (absent) to 4 (severe).

Statistical Analysis

Stata® 12 was employed for data analysis. Continuous variables are described as mean ± standard deviation. To compare the animal weight between the beginning and the end of the study in phase 3, a two-sided Student's t test was employed. A P value of less than 0.05 was considered significant.


Results were considered as operative findings, general response to surgery, gross examination and microscopic examination.

Operative Findings

Operating duration was 42.5 ± 5.75 min. Operations were uneventful. Mean BLPP with one cuff closed was 1091 ± 30.6 cmH2O. With two or three cuffs closed at the same time, BLPP was 1163.5 ± 71.4 and 1244.1 ± 99 cmH2O respectively. The sphincter worked correctly in all animals during the operation and in the postoperative period except for one animal in the third phase. In this sheep one cuff was seen to have failed at explantation. This failure had been recorded by the computer on the fourth day after implantation. The computer registered no other failure in any component in any other sheep.

General Response to Surgery

No death or clinical infection occurred in animals in the period considered. No significant animal discomfort or pain was noted. Sheep voided only when the device was switched off. All vital parameters and urine volumes were normal, adjusted for animal's weight and sex. In phase 3, sheep weight before implantation was 88.9 ± 8.7 kg. Health status after implantation was considered good since body-weight at the end of the experiments was stable at 91.4 ± 7.9 kg (P = 0.56) (Table 1).

Table 1. Animals' weight at the beginning (T0) and at the end (T1) of the study.
AnimalWeight T0 (kg)Weight T1 (kg)LymphocytesPmnNecrosisEncapsulationCell/tissue degenerationImmunoreactivity of macrophages
  1. Pmn, polymorphonuclear leucocyte. Histological characteristics of the tissue are given in a score from 0 to 4 in which 0 means absent, 1 slight, 2 moderate, 3 marked, 4 severe.

Gross Examination

In all explants, the cuffs and the control unit were correctly positioned (Fig. 5). No degradation of the device was observed. Urethra surrounded by a cuff had no sign of erosion, infection or stricture and seemed to be as elastic as the control without evidence of tissue stiffness.

Figure 5.

Macroscopically explanted cuffs are correctly positioned. The urethra shows no sign of infection or atrophy.

Microscopic Examination

In the phase 2 study there was no sign of infection, inflammation, tissue necrosis or erosion at any time point. In the phase 3 study there was no significant difference in tissue architecture of the urethra within the cuff compared with the urethra at any other site, local or remote (Fig. 6). Table 1 summarizes the histological findings. Overall the local tissue response was considered good. Marked cellular damage, necrosis and moderate polymorphonuclear presence were seen in one sheep from the phase 3 study. In another sheep a slight degree of necrosis and lymphocyte infiltration was observed. Otherwise there was no significant lymphocyte or polymorphonuclear cell infiltration, necrosis or cell degeneration. The implanted material elicited a chronic foreign body reaction in the peripheral tissues, characterized by the presence of a degree of encapsulation in all specimens. In one sheep a necrotic reaction was observed at the interface between the urethra and the annular polymer part of the device. Immunohistochemistry showed the absence of macrophages in the samples.

Figure 6.

Cross-sections of urethra surrounded by a cuff demonstrate the normal structure compared with control urethra.


Since Foley's first attempt in 1947, various devices and techniques have been developed to replace sphincter activity. Pneumatic, magnetic and hydraulic mechanisms have all been tried [17-20]. Recently, novel slings have been developed and the first results seem promising [7-12]. However, for refractory severe stress urinary incontinence, the AMS 800 is still considered the gold standard [1]. Despite the latest improvements and good clinical outcomes, the revision rate is still high. AMS 800 failures can be divided into two categories, mechanical and non-mechanical, with a variable but roughly equal incidence [3]. Mechanical failures include leakage and malfunction of the device itself. Non-mechanical failures include cuff erosion, infection and urethral ‘atrophy’ affecting the compressed area of the urethra. Since the compressed area of the urethra is always the same the vascularity of this area is potentially compromised by a hydraulic device and urethral ischaemia can occur as a consequence. Although the urethra is commonly considered to be a passive tube through which urine flows, it has been shown that the area of the urethra within the AUS cuff that constitutes the continence mechanism in patients with an AUS is at least partly dependent on the preservation of urethral vascularity [13].

Attempts have been made to produce a better hydraulic device. In one particular study a self-sealing port was included in the pump to adapt the occlusion pressure to each patient and a stress relief mechanism was added to minimize closure pressure. Although the effectiveness of the device was demonstrated, no conclusion regarding long-term outcomes could be made since only three patients were still in the trial after 12 months and no comparison group with the standard AMS 800 was used in this preliminary clinical investigation [14].

The ideal AUS should preserve urethral vascularity and be individually adjustable postoperatively. The emAUS was developed with these goals in mind. Furthermore the procedure is simple because dissection is limited and there are no connections to be made. The significant advantage of the emAUS for human application is its adjustability after implantation. The remote control allows the cuff closure pressure to be adjusted exactly to a patient's needs. As a result the lowest closure pressure to give continence will be applied and this will contribute to the long-term preservation of urethral vascularization.

Nitinol was chosen for the contractile unit because of its excellent performance, its low energy consumption and because it can be activated repeatedly if not indefinitely without material damage. Energy consumption of a nitinol device is less than with other materials. Nevertheless the power supply of the device remains a problem. The emAUS used in this animal study was powered by an external energy source. For clinical practice we have developed a remote external transmitter that will operate an implanted receiver to control the contractile unit, rather than the transcutaneous external board used in these experiments. Currently it will work for 1 week without recharging. We plan to supply power by transcutaneous recharging and this is technically feasible but we continue to investigate other potential solutions.

The emAUS is clearly effective in producing continence. Mean BLPP measured during the operations was much higher than BLPP in continent patients in physiological conditions. Indeed the force used to close each cuff could probably be reduced considerably in clinical practice from the 0.7 N used in these experiments.

The major finding in this study was the good tissue response to implantation. The key concept is the use of electronics to produce sequential compression of alternating segments of the urethra, thereby giving continence, with preservation of urethral vascularity by avoiding constant compression of alternating segments since this avoids permanent compression of any single segment of the urethra. That this is actually the case is shown by the absence of structural damage on microscopy.

Clearly this approach to urethral compression requires at least two cuffs. More cuffs are possible but additional cuffs mean that more electromotor and mechanical parts are required. This in turn means that the device will be bigger and heavier and more space will therefore be required to implant it around the urethra. To overcome these issues we settled on a two-cuff emAUS in phase 3.

A comparison with other experimental studies is difficult because we have only been able to identify two studies, both in different animal models. The original AMS device was tested in an animal study in eight mongrel dogs [15]. Explantation was performed between 4 and 26 weeks. There was no evidence of tissue necrosis, urethral stricture formation or cuff migration and the authors found the same foreign body reaction with encapsulation of the prostheses as we found in the present study. The second study, of a magnetically operated sphincter working in an entirely different way, was stopped after 6 weeks because continuous compression on the skin caused significantly reduced blood flow and pressure ulcers in miniature pigs [16].

One potential criticism of these studies is that the sheep were not actually incontinent but as they were rendered unable to void with the device active and could only void when the device was switched off it was clearly capable of producing continence in incontinent subjects. Another criticism might be the absence of a sham group or a comparative group using the AMS 800. However, as there was no adverse response to the device in the phase 2 study we felt it was unnecessary to include a sham group in the phase 3 study when we would otherwise have considered it; and we felt a comparative study was unnecessary in this pre-clinical phase when we were trying to demonstrate a proof of concept.

Finally, of course, these are only short-term data with a maximum length of 3 months' implantation. However, given that there were no adverse histological consequences whatever at this stage we feel optimistic as we proceed to the next stage of development of this device.

We have developed a new emAUS, based on electronics and implanted contractile units. This electromechanical mechanism provides a sequential alternating compression to close the urethra in successive segments. In this proof of concept, we have demonstrated its effectiveness. Histological findings show that the mechanism can avoid urethral atrophy and ischaemia by the preservation of urethral vascularity. Further technical developments in power supply and energy consumption as well as long-term in vivo tolerance data are required before human use could be considered.

Conflict of Interest

Anthony Mundy and Robert Dahlem are scientific advisors for MyoPowers SA. Piergiorgio Tozzi is one of the founders of MyoPowers SA.


MyoPowers SA, CTI project No 9712.2 PFLS-LS and the University Hospital Lausanne supported this study. The results of the first and second phase were presented at the European Association of Urology congress in Barcelona in March 2010. The abstract won the Third Prize for the Best Abstract (Non-Oncology) of the congress and the Third Prize for the Best Abstract by a resident. Third phase results were presented at the American Urological Association congress in Washington in May 2011.