Implantation of SphinKeeperTM: a new artificial anal sphincter

Introduction A new artificial anal sphincter, SphinKeeperTM, was devised with the aim to treat fecal incontinence (FI) by implanting specifically designed self-expandable prostheses into the intersphincteric space. Preliminary data concerning the procedure feasibility and prosthesis localization at 3 months are presented. Methods SphinKeeperTM prostheses in the native state are dehydrated, thin, solid cylinder (length 29 mm, diameter 3 mm), changing their state (shorter—length 23 mm, thicker—diameter 7 mm—and softer, with shape memory) within 48 h of contact with fluids. In this study, 10 prostheses were implanted in each patient with FI under local anesthesia and under endoanal ultrasound (EAUS) guidance, into the upper-middle intersphincteric space of the anal canal by a specifically designed delivery system. EAUS was used postoperatively to assess prostheses dislocation. Results Ten patients (5 females; median age 58 years, range 20–75) were enrolled and treated with SphinKeeperTM implantation. Median duration of procedure (performed by endoanal ultrasound guidance) was 40 min (range 30–45). Neither intraoperative nor postoperative complications were reported after a 3-month follow-up. In one patient, a partial dislocation of a single prosthesis was documented by EAUS, causing anal discomfort which resolved after 1 week. Conclusion SphinKeeperTM can be safely implanted in patients with FI of different etiology. Implantation was well tolerated with no dislodgment of implants at 3-month follow-up.


Introduction
Multifactorial etiology of fecal incontinence (FI) has a significant impact on the choice of management [1][2][3]. Sphincter lesions are considered to be the main cause of FI, particularly in female patients, but frequently the dysfunction occurs also in subjects with intact sphincters. In other cases, neuropathy (either peripheral or central) plays the pivotal role, causing sensory-motor alterations [2][3][4][5][6][7]. Also the severity of FI can be variable, ranging from soiling, seepage and incontinence to gas (commonly defined as ''minor incontinence'') to incontinence to liquid and solid stools (defined as ''major incontinence''). Despite the numerous modalities of treatment available, the therapeutic efficacy is still suboptimal for all of them. In fact, ''conservative'' therapies such as biofeedback have high failure rates while the success of ''minimally invasive'' procedures such as injectable bulking agents, radiofrequency, tibial and sacral nerve stimulation or ''aggressive'' procedures such as anal sphincteroplasty, graciloplasty, artificial bowel sphincter and magnetic sphincter range from partial success to complete failure [8].
A new artificial anal sphincter, SphinKeeper TM , was devised with the aim to surround the anal canal with specifically designed self-expandable prostheses placed into the intersphincteric space. Here we report the safety data and the short term ultrasound assessment after implantation of this device in patients with FI. & C. Ratto carloratto@tiscali.it

Materials and methods
SphinKeeper TM prostheses (THD SpA, Correggio, Italy) are made of hyexpan (polyacrylonitrile), which was previously used for the same purpose in the Gatekeeper TM (THD SpA, Correggio, Italy) [9,10]. SphinKeeper TM prostheses in the dehydrated state are, thin, solid cylinder (length 29 mm, diameter 3 mm), changing their size (shorter-length 23 mm, thicker-diameter 7 mm) and physical properties (softer, with shape memory) within 48 h of contact with fluids ( Fig. 1).
Patients were included in this study according to the following criteria: older than 18 years and younger than 80 years, FI onset at least 6 months before the implant, a number of FI episodes more than once a week and willingness to perform baseline and follow-up schedule evaluations and to sign an informed consent. Exclusion criteria were: malignancies under treatment, rectal bleeding of unknown origin, chronic diarrhea unresponsive to medical treatment, inflammatory bowel disease unresponsive to medical treatment, acute anorectal sepsis, concomitant rectal prolapse, obstructive defecation syndrome, neurological disease, coagulation disorder. At baseline, patients were evaluated on the base of detailed medical history and physical examination; when indicated, a colonoscopy was performed. Particular attention was paid to investigate previous surgery, trauma and/or local radiotherapy, congenital anorectal malformations, comorbidity, FI symptoms, characteristic and diary, ability to defer defecation, need to wear pads and/or taking constipating drugs, Cleveland Clinic Fecal Incontinence Score (CCFIS), Vaizey score, AMS score [11][12][13]. Health status and quality of life were evaluated with SF-36 and FIQL, respectively [14,15]. Anorectal manometry (ARM) and endoanal ultrasound (EAUS) were performed to assess the anorectal function and morphology.
The aim of this study was to record of intra-postoperative adverse events in order to obtain data about technical feasibility and safety. Intra-and postoperative complications were prospectively collected, in particular looking at clinical and/or EAUS evidence of bleeding at/from the sites of implantation, anorectal sepsis (anorectal abscess and/or fistula), signs of local and systemic inflammation, pain, duration of analgesics therapy and urinary retention. Prosthesis were checked with EAUS at 1 week, 1 month and 3 months for dislocation.

Operative procedure
Patients underwent standard bowel preparation with two 120 ml docusate sodium enemas, one the night before and one few hours before the operation. Antibiotics prophylaxis was given as 1 g of i.v. Cefazolin and 500 mg of i.v. Metronidazole. All implant procedures were performed with the patient in lithotomy position, under local anesthesia using a Mepivacaine perineal block. Skin was prepped with povidone iodine solution. The entire procedure was carried out under EAUS guidance. The delivery system (Fig. 2) was loaded before each insertion. Before insertion an activating button was pushed for 5 s resulting in the extrusion of a guiding cannula. Ten 2-mm perianal skin incisions were made 2 cm from the anal margin, equidistant each other. Under digital guidance, the introducer was inserted into the intersphincteric space through a short subcutaneous tunnel and pushed up to reach the upper part of the anal canal. EAUS confirmed the correct position of the cannula into the intersphincteric space and the tip of the introducer at the level of the upper third of the anal canal. Once in place the gun was fired resulting in a retraction of the cannula and deployment of the prosthesis. When the proximal part of the prosthesis reaches the tip of the cannula, the whole cannula completely retracts inside the delivery system and the prosthesis is released in place in the desired position within the intersphincteric space. The delivery device was withdrawn. EAUS is used to confirm the position of the prosthesis, which appears as an hyperechoic dot with a hypoechoic shadow behind it. Three-dimensional EAUS was used by our group and showed a continuous hyperechoic line. The same procedure was repeated for all 10 prostheses around the entire circumference of the internal anal sphincter (Fig. 3). At the end of the procedure, a final 3D-EAUS imaging was carried out as final assessment (Fig. 4). Finally, skin wounds were sutured with resorbable material. Repeated EAUS at 1-week documented the final dimensions of the prosthesis (Fig. 4). All patients were strongly advised to observe bed rest or slowly move out of bed to chair for 48 hours to minimize early prosthesis dislocation. Lidocaine gel and systemic painkillers (ibuprofen and tramadol) were prescribed as needed for postoperative pain.

Postoperative evaluation and follow-up
Clinical evaluation (comprehensive of FI symptoms, continence diary, FI severity scores-CCFIS, Vaizey and AMS, FIQL and SF-36 questionnaires, physical examination and EAUS) was scheduled at 1 week, 1 month and 3 months after SphinKeeper implant. All patients were informed about the importance of avoiding constipation and hard stool; a diet rich in water and fibers, and oral assumption or stool softeners were prescribed during the first postoperative month.

Results
In the period between July 2014 and April 2015, 10 consecutive patients (5 men; 5 females) with a median age of 58 years (range 20-75), who met the inclusion criteria, were enrolled in the present pilot study and treated with SphinKeeper TM implantation. Table 1 summarizes the baseline patients' characteristics, including data from clinical history, FI severity scores, previous management of FI, health status and quality of life of the enrolled patients.
Minimum duration of FI was 3 years, while the median duration was 9 years. Four patients referred passive FI, 4 urge FI and 3 mixed FI. Seven patients had undergone perineal surgical procedures, and 3 of them previous surgery for FI. One patient had undergone pelvic radiotherapy for endometrial cancer. Nine patients used pads daily. Baseline ultrasonography and manometry data are reported, respectively, in Tables 2 and 3.
Median operative time for SphinKeeper TM implantation under EAUS guidance was 40 min (range 30-45). Neither intraoperative nor early postoperative complications were reported during the hospital stay.
At 1-week EAUS evaluation, all the prostheses reached their final size, even those implanted in scar tissue (Fig. 5). At one week, one patient complained of anal discomfort that, at digital examination and EAUS evaluation, was attributed to a 1 cm distal dislocation of a single prosthesis within the intersphincteric space, which was treated by local and systemic painkillers (lidocaine 2 % gel and paracetamole 1 gr tablets) until the resolution of symptoms one week from the onset.
All patients completed the follow-up evaluation at 1 week, 1 month and 3 months. There was no postoperative morbidity, in particular in terms of obstruction to the

Discussion
Our study shows that SphinKeeper TM implantation is not followed by septic or adverse reactions, does not result in short-term dislodgement and is well tolerated. This is promising, given the size, number, site of deployment and elastic property of the prostheses, which are novel and discussed below. Functional results and quality of life are not the scope of this report given the short follow-up of 3 months.
SphinKeeper TM is the result of multiple innovations in the treatment of FI, concerning both the device and the implantation procedure therefore deserving the title definition of ''new artificial anal sphincter.'' In terms of biomechanics, SphinKeeper TM prostheses are made of a material which has already been clinically tested [9,10] and has the ability to rapidly (within 2 days) self expand as well as shape memory. Because of these characteristics, the prostheses, even if inert have the potential to change in shape with stool passage through the anal canal (because soft and compressible), and regain shape and size afterwards. This mechanism of action can be particularly useful when the prostheses are placed into the upper and middle anal canal, where, physiologically, the recto-anal inhibitory reflex is elicited, starting the cascade of defecation events. On the other hand, no obstruction developed during the stool passage, as demonstrated in the present study.
Prostheses fit well into the anal canal. They are long enough (23 mm in the final length) to reconstitute the normal anal canal length and wide enough (7 mm in the final diameter) to ensure a significant filling ability. Moreover, the documented expansion of SphinKeeper prostheses also within the scar tissue could give the  opportunity to treat also patients with sphincter defects. These aspects should be further investigated in large number of patients.
The SphinKeeper TM procedure provides the implantation of 8-12 prostheses (always 10 in the present study). The prosthesis is made of a material which has been prospectively tested in the setting of fecal incontinence [9,10], and has the advantage of being easily visualized by ultrasound. This allows the operator to choose accurately the implant site before the prosthesis release and, then, activate the mechanism of delivery. The insertion and deployment of each prosthesis takes only a few seconds.
The choice to implant the prostheses into the intersphincteric space of the anal canal plays a key role. This location potentially avoids extrusion or migration of prostheses (differently than if implanted into the submucosa). Moreover, thanks to the rapid increase of their volume, the prostheses are unlikely to move after deployment. The bed rest which was recommended in this pilot study is aimed at decreasing the event of an early prosthesis dislocation (which occurred in one out of 100 prostheses implanted in this study). As far as postoperative complications SphinKeeper TM implant was very safe in this study; in fact, no acute sepsis at the site of implantation and around the prostheses was documented within 90-day period. No External anal sphincter lesion 4 Internal anal inhomogeneity 5 External anal inhomogeneity 5  patient had long-lasting symptoms (including anorectal pain and discomfort) directly or indirectly related to the implanted prostheses. This confirmed what was already observed in a large population submitted to another type of implantation of this material into the anal canal [9,10]. Specifically in the case of SphinKeeper TM , a high number of prostheses are implanted reaching a very high final volume of implanted material (8650 mm 3 , approximately 480 % increase in size of the native sphincter), surrounding the anal canal and playing the role of an ''additional'' sphincter. The changes in the sphincter anatomy confirmed by imaging are expected to play a physiologic role. The hypothetical positive interaction between the Sphinkeeper TM prostheses and the adjacent IAS and EAS is interesting. Under physiological conditions, central input (neural drive to the muscle) and muscle length (microscopically, sarcomere length) are the key determinants of tension and force generated by the skeletal muscles [16]. It has been well demonstrated that EAS muscle operates at a short sarcomere length in both rabbits and humans; in other words, increasing its in vivo length increases its contraction [17][18][19][20][21]. Our hypothesis is that the large volume Sphinkeeper TM implants, placed between EAS and IAS (pushing the EAS outwards and the IAS inwards), may increase the muscle fibers length and, therefore increase their contractility. Further studies will definitely elucidate this mechanism of action. These properties make of SphinKeeper TM an attractive alternative to the ''external'' artificial anal sphincters (ABS, i.e., artificial bowel sphincter, dynamic graciloplasty, slings, magnetic anal sphincter). In fact, in the case of ABS and dynamic graciloplasty, only the release of the closure system, operated by the patient, can permit the defecation, while, in the case of anal slings and magnetic anal sphincter, the pressure of the rectal content allows the anal canal opening. We hypothesisze that, SphinKeeper TM , as embedded into the anal canal, may improve sphincter contractility by increasing sarcomere length as well as increase the length of the anal canal and provide a powerful ''bulking effect.'' Following our hypothesis, in patients with loose, patulous, funnel-like or keyhole-shaped anal canal, SphinKeeper TM could offer the opportunity to reconstitute the cylindrical shape of the anal canal, while in patients with sphincteric lesions it could reinforce the area of scarring improving the contribution to the continence by the remaining intact sphincters. Finally, not insignificant could be the role played by SphinKeeper TM as adjunctive therapy in patients with incomplete resolution of symptoms after other procedures for FI. In conclusion, implantation of the SphinKeeper TM is feasible in patients with different types of FI and does not give rise to major complications or dislodgement. If proved effective it may be a viable, minimally invasive first line approach in patients suffering from FI.

Compliance with ethical standards
Conflict of interest The authors declare that they have no conflict of interest.
Ethical approval All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Informed consent Informed consent was obtained from all individual participants included in the study.