FI management is, above all, based on regularizing stool frequency and consistency. Use of a bowel diary combined with a food diary may help identify food that aggravates stool leakage, such as caffeine, fruit and vegetables, alcohol, or spicy food[10]. Cutting out these foods may alleviate FI for some patients even though definitive evidence is lacking[11]. A high fibre diet, or the consumption of mucilage can, in some cases, improves faecal continence by improving stool consistency, especially for patients with liquid stools or with associated constipation. In a randomized controlled study, Bliss et al[12] showed a 50% reduction in the number of FI episodes per week after daily consumption of Psyllium compared to placebo.

Several different medications can be used to regularize stool frequency, improve stool consistency or enhance anal canal tone. Anti-motility drugs (loperamide, codeine) have been scientifically proved to be efficient for patients suffering from FI[13]. These molecules seem to be the most useful in cases of loose of liquid stools by reducing stool frequency and thus the number of incontinence episodes. Indeed, in a double-blind cross-over trial comparing loperamide with placebo, about 60% of the patients had less incontinence episodes with loperamide than with placebo[14]. Their efficacy has not been proved in the cases of normal or hard stools.

Stool-bulking agents, such as mucilage, are also useful in management of FI with loose stool even if these practice is not supported by strong scientific studies. These synthetic fibres improve stool consistency by increasing water absorption by stools. Even though their efficacy has not been rigorously proved scientifically, mucilages are still recommended in clinical practice[15], particularly combined with other conservative treatments[16].

Ion-exchange resins may also be tried in FI management. Cholestyramine, in particular, has been shown to improve stool frequency and consistency in patients who have an FI in combination with diarrhea[17].

Several pharmaceutical substances have been studied in FI that aim to enhance anal canal tone and thus reduce the number of incontinence episodes. Topical agents such as phenylephrine[18-20], or oral treatments such as diazepam, amitriptyline or sodium valproate[21-23] have shown some efficacy in this indication, sometimes up to 100% of treated patients. However, given the small number of studies and the potentially disabling adverse effects, these treatments are not recommended in clinical practice.

Even though some data suggest a potential efficacy of hormone replacement therapy on FI in postmenopausal women, about 65% of treated patients[24], it is not currently recommended due to insufficient scientific proof[15].

Finally, in cases where FI is associated with rectal emptying disorders, help with rectal emptying is an essential point in incontinence management. Suppositories and rectal irrigation, combined with oral laxatives if necessary, all contribute towards eliminating faecal impaction, thus allowing a decrease of 35% in the number of FI episodes and of 42% in faecal soiling especially in geriatric patients[25].

If the first line medical treatment fails, perineal rehabilitation strategies can be offered. These physiotherapy techniques are based on re-training faecal continence and target both anal sphincters and abdominal muscles. Anorectal manometry is used to determine the most adequate therapy.

These techniques mainly involve pelvic floor exercises, anal electrostimulation and biofeedback. Pelvic floor exercises are muscle strengthening exercises, especially of the pubococcygeus using Kegel exercises[26]. Anal electrostimulation uses surface electrodes on the perineum which either stimulate muscular contraction directly or indirectly via stimulation of peripheral nerves[27]. Biofeedback therapy helps to increase voluntary contraction of the external anal sphincter, but also to synchronize the different perineal muscles in response to a rectal stimulus in order to maintain continence[28]. This technique uses instruments capable of monitoring sphincter contractions and thus helps with training.

Perineal rehabilitation strategies have shown heterogeneous efficacy on FI depending on the study. Despite anal electrostimulation having been shown to be beneficial in this indication[27], its use alone is not recommended[15]. Biofeedback therapy seems to be the most widely used and the most efficient, with success rates between 50% and 90% that were maintained up to 24 mo[29,30]. Even though there are discordant results in scientific literature[31], these perineal rehabilitation techniques are still recommended in second line for FI management.

PTNS appears to be a simple technique to use, it is non invasive and not costly. Two methods of stimulation exist: Percutaneous, using needle-electrodes, and transcutaneous using adhesive surface electrodes. Two electrodes are placed on the posterior tibial nerve pathway, and linked to a stimulator that can be controlled by the patient. The mechanism involved in FI treatment remains poorly understood but certainly involves afferences and somato-sympathetic reflexes. Even though there are only a few studies published with relevant different results for the two methods of stimulation[32,33], this technique reduced FI episodes for 63% to 82% of patients treated, with a follow-up of 1 to 30 mo[34]. To this day, there is no consensus concerning treatment duration, stimulation frequency/rythm or need for repeating treatment. However, this technique may be recommended for patients who do not respond to the other non-invasive techniques or suffering from FI without transit disorders[15].

TAI is currently recommended in second line management, after dietary measures and first-line medical treatment, in patients suffering from chronic neurological diseases[15]. The aim is to empty the colon of the maximum of faecal matter using regular irrigation, optimized using an inflatable rectal balloon catheter to make the system watertight. This method improves digestive symptoms, including FI, in between 40% and 75% of patients suffering from chronic neurological diseases[35-37]. It also helps to significantly enhance the patients’ QOL, by significantly increasing their independence, and seems to decrease the risk of urinary infections[38]. However, this treatment can only work if the patient and their family are committed.

Moreover, this device may also be used in non-neurological patients as demonstrated by recent studies in patients suffering from anterior resection syndrome. For example, in the study by Rosen et al[39] after a mean follow-up of 29 mo, TAI allowed a significant improvement in the number of stools, the Cleveland Incontinence Score, and in QOL. A study by Koch et al[40] using a device from another manufacturer in thirty patients, showed a complete improvement in 57% of patients.

Anal plugs come in different sizes and are made from different substances, they are easy to use and can be of great help on a daily basis. Tolerating anal plugs can be a problem but can be improved depending on the type of plug used[41,42]. Anal plugs have shown to be efficient on FI and can be used as a supplementary treatment to other therapies. Indeed, when patients tolerate them, up to 65% of patients reported the absence of soiling episodes[43].

Minimally invasive treatment: When conservative therapy has not been sufficient, some minimally invasive methods can be proposed to improve faecal continence. Currently, this type of therapy is mainly represented by sacral neuromodulation (SNM). Other techniques exist but they are more confidential, such as antegrade colon irrigation, anal radiofrequency treatment or intrasphincteric injections.

SNM is proposed to patients suffering from at least one FI episode a week and for whom conservative treatment, combining hygiene/dietary control and biofeedback, has failed[15]. The mechanism of action remains poorly understood to this day[44], although it seems to involve several mechanisms at different levels. Indeed, it could act on continence via somato-sympathetic medullary reflexes[45], but also via the activation or inhibition of areas of the brain responsible for continence[46]. This cerebral mechanism was also discussed in a recent study that showed a persistent efficacy of SNM even after stimulation had been stopped, suggesting that a long period of SNM could bring about a certain neuroplasticity[47].

This treatment stimulates the sacral nerves on a permanent basis via an electrode implanted in contact with the nerve where it exits the sacral foramen. The device is set up in two stages. The first stage, called peripheral nerve evaluation (PNE), is a test period. For 2 to 3 wk, the electrode is implanted, the most often in contact with the S3 root[48,49], and linked to an external stimulator. Stimulation parameters can therefore be modified to obtain sufficient efficacy on FI. The second stage, involving definitive implantation of the pulse generator under the skin, is only carried out if the patient has a 50% reduction of FI episodes with PNE[15].

On the whole, in a recent literary review, SNM seems to be efficient in approximately 60% of patients suffering from FI, for whom conservative therapy failed. The therapeutic effect persists over time, even though a 10% reduction in efficacy was noted during the first 5 years[50]. At the same time, 70% to 80% of patients who had this treatment declared that it had improved their QOL[51-53].

Indications for SNM are numerous. It can be recommended in cases of idiopathic FI, or FI linked to post-obstetric perineal injuries, such as anal sphincter tears or stretch-induced neuropathy[54-56]. It is also recommended in cases of FI of neurological origin, either central or peripheral[57-59]. Patients with systemic sclerosis suffering from FI can also benefit from SNM, although recent data have shown a lack of efficacy for this indication[60]. Finally, patients suffering from double incontinence, urinary and faecal, seem to respond to SNM[61,62]. Other aetiologies of FI are not currently considered to be indications for SNM[15], even though there is some data, particularly for Crohn’s disease[63].

Several studies have tried to determine predictive factors for SNM efficacy on FI. Factors such as age, stool consistency, symptom duration, pre-therapeutic manometric data, or obtaining a motor or sensory response at a low stimulation threshold have all been suggested to be predictive of a good or bad response to SNM. However, despite the fact that data is sometimes contradictory and that recent studies have not been able to identify any significant predictive factors[64], SNM should be considered to be a therapeutic option for all patients suffering from FI, except when contraindicated[65].

Based on the same principal as TAI, antegrade irrigation aims to restore continence by keeping the colon empty. The surgical procedure creates a caecostomy allowing patients to perform colon irrigation themselves.

With this technique 80% to 90% of patients reported pseudo continence, with, however, some complications that sometimes led to explantation of the device[66,67]. This method has also been shown to be efficient when combined with an artificial urinary sphincter in patients suffering from double incontinence[68].

More recently a technique of percutaneous endoscopic caecostomy has been proposed with interesting results but further studies are necessary to assess its success rate in the treatment of FI[69].

Temperature-controlled radiofrequency energy delivery to the anal canal (SECCA procedure) is an endoscopic technique mainly used for passive FI, with a success rate of close to 70%[70]. It is based on forming retractile fibrosis, with deposition of collagen and shrinkage in the internal anal sphincter. However, published results are sometimes contradictory[71] and in the long term its efficacy seems to rapidly decrease[72].

In order to increase anal resting pressure, several different bulking agents, injected either into the submucosal or intersphincteric space, have been tested on FI[73]. Efficacy varied depending on the product tested[74-76], but only two studies compared bulking agent injections with sham injections. Dextranomer microspheres in stabilised hyaluronic acid (NASHA Dx) was the only product that showed a significant difference vs placebo. Indeed, a 50% or more reduction of the number of FI episodes was observed in 52% of patients who received NASHA Dx and 31% of patients who received placebo[77]. A large number of adverse effects were reported, but most of them were not serious. On the other hand, despite a significant improvement on continence, injections with silicone elastomers showed no significant difference vs sham injections, underlining the noticeable placebo effect of this procedure[78]. Finally, in a randomized clinical trial studying an injectable silicone biomaterial (PTP™), the success rate was 69% with endoanal ultrasound guidance and 40% without[79]. So, this technique seems to give better results when the injections are performed under endoanal ultrasound guidance.

Surgical treatment: Several different surgical solutions can be offered to patients with severe FI[80]. The most frequently used technique is sphincteroplasty. Given the good results obtained with conservative treatment and SNM, indications for surgical treatment have become relatively rare.

Sphincter repair is indicated in patients with symptomatic FI associated with external anal sphincter damage[15]. It aims to restore, at least partially, the anatomical barrier necessary for faecal continence and is recommended in priority in patients with recent sphincter injury that does not exceed half the circumference[81].

Several different surgical techniques have been described, notably direct repair and overlapping sphincter repair, but no significant difference was demonstrated[82,83]. Performing a temporary protective colostomy did not show any probing results either[84].

Short-term efficacy on FI is close to 70%[85]. This efficacy tends however, to decrease with time[86]. Despite this, patient satisfaction and QOL remained high even on a long-term basis[87]. Furthermore, in cases of persistent and debilitating FI, it is possible to repeat surgical repair on the external anal sphincter[88].

Techniques of muscle transposition aim to replace anal sphincters, especially in cases where sphincter damage is too severe. Several techniques have been described but graciloplasty, dynamic or not, is the most studied and used technique.

The success rate of graciloplasty is between 60% and 90%[89-93]. Originally data seemed to suggest that dynamic graciloplasty was the most efficient, however a recent study reported an identical success rate for both techniques[94].

Despite a high rate of complications, between 35% and 75%, (infections, wound healing, pain, stoma problems, constipation) and the common necessity for further surgical procedures[95-98], these methods of muscle transposition remain efficient on anal incontinence, with a success rate which persists over time[99,100]. Furthermore, patients who have had graciloplasty report an improved QOL[101].

However, in clinical practice, the indication of these techniques remains limited[15].

An artificial anal sphincter, whatever the model, is made up of a band around the anal canal, a pump placed in the external genital organs, and a pressure-regulated balloon[102]. It is especially indicated in case of severe sphincter damage and/or severe local neurological lesions.

In a recent meta-analysis, the technique’s success rate on FI was 75% in the first three years, followed by a progressive decrease since it was only 55% after 5 years of follow-up[103]. Other studies have shown more disappointing long-term results, with only 3 out of 25 patients in the study by Altomare et al[104] and 4 out of 21 patients in the study by Darnis et al[105] showing a good functional result after 3 years of follow-up.

Along with these contrasting results, morbidity of the technique is far from being negligible with a rate of explantation of the device between 24% and 39% depending on the length of follow-up[103]. The majority of complications are infections, perianal pain, and, on a long-term basis, problems with rectal emptying. In a recent case study, all 21 patients treated had at least one complication during the 38-mo follow-up[105].

This high rate of complications combined with moderate efficacy limits the use of this technique in common practice.

Often reserved for patients for whom all the other treatments have failed, colostomy can be indicated to treat severe FI[106]. Comparison of QOL between patients with a colostomy and those suffering from FI, showed a higher social function and an improvement of the coping, embarrassment, lifestyle scales and depression scales in the colostomy group[107]. In the same way, after colostomy, patients who initially suffered from FI granted a high level of satisfaction, in spite of initial apprehension and some complications reported (bleeding, parastomal hernia, mucus leakage)[108]. Therefore, this technique, used in strategic FI management, should not be excluded as a failure since in the end it improves patient QOL and grants them more independence.

Alpha1-adrenoceptors have been tested for years since they are known to induce a contractile response of the human internal anal sphincter[109]. However because of their poor clinical tolerance (modifications of cardiovascular parameters), their use didn’t initially spread in clinical practice. More recently, since NRL001 has a dose-dependent effect, the concentration was modified to improve tolerance. However, to date, only phase-I studies are available[110-113]. We are currently waiting for the future results of the “Libertas” study that is a multicentre Phase II, double-blind, randomised, placebo-controlled, parallel group. It will evaluate the efficacy, safety and tolerability of locally applied NRL001 in patients with FI[114].

The use of stem cell can be done in two main different ways: Direct injections in the sphincter or peri-anal implantation of a bioengineered sphincter. Stem cell involved may be mesenchymal stem cell or muscle-derived stem cell. For the moment, most of the studies have been done in animal models. The first study reporting that autologous muscle-derived stem cell grafts (in a rat model) may be a new tool for improving anal sphincter function was published in 2007[115]. Then, other studies have demonstrated the feasibility of the use of stem cells in a rat model, with myoblast[116] and with mesenchymal stem cells[117,118]. Another way to use stem cell is the perinanal implantation of a bioengineered sphincter. In an original study, isolated human internal anal sphincter circular smooth muscle cells and human enteric neuronal progenitor cells were use to construct a bioengineered internal anal sphincter. After maturation, it was implanted in the perianal region of athymic rats and retrieved from the animal after 4 wk. The implantation was well tolerated in all animals and there were no postoperative complications. Normal stooling was observed during the implantation period. After the 4 wk period, it was observed that implanted bioengineered sphincters were adherent to the perirectal rat tissue and appeared healthy and pink and immunohistochemical data showed neovascularization[119]. A particular interest of this model is that the bioengineered IAS may overcome the problems of foreign bodies in the perineum since it is purely made from autologous tissue. Further studies will be necessary to confirm this concept.

To date, only two works describe the use of stem cell in humans. The first one was published in 2010 and included 10 women suffering from FI due to obstetric anal sphincter injury. In this pilot study, autologous myoblasts were cultured from a pectoralis muscle biopsy, harvested, and then injected into the external anal sphincter defect under ultrasound guidance. At 12 mo a significant improvement in the Wexner score and in the QOL were observed. The anal squeeze pressures did rise significantly at 1 mo and 6 mo post-injection without persistence at 12 mo. The procedure was well tolerated and no adverse events were observed[120]. The second case, published in 2013, has reported one case of injection of myoblast also obtained from a sample of the quadriceps muscle. The patient was a 20-year old male with FI due to an old external anal sphincter rupture in a road accident[121]. Although this work also showed encouraging results, data are still scarce and these two studies essentially demonstrate the feasibility in humans. Further studies will be needed to confirm the feasibility of stem cell injection in FI and to assess the potential long term success of this method that is for the moment limited to experimented centres.

The Fenix^® magnetic anal sphincter operates on the principle of a reverse stent system; it is composed of a magnetic bead that creates a negative pressure around the tube it encircles. The device is made of magnetic balls tied together by titanium threads. Various magnetic sphincter lengths are available to accommodate the individual variations of anal circumference. This sphincter is designed to enhance the function of the anal sphincters without causing obstruction. It is functional immediately after implantation. During stool evacuation, the patient strains in a physiological way to create a sufficient force to separate the magnetic balls and to open the anal canal permitting thus the passage of stools.

The first feasibility study was published in 2010 and included results for 14 patients from[122]. The authors described a simple technique with a low morbidity (mainly represented by surgical site infections). Compared to the conventional artificial sphincter, the operative time and duration of hospitalization were much shorter. Since this first study, the results of the Fenix^® magnetic anal sphincter device on FI have been reported in the literature, but few studies are currently available. In 2012, Wong et al[123] reported results from a single-center non-randomized study showing that the magnetic anal sphincter was as effective as sacral nerve stimulation in improving symptoms and QOL in patients suffering from faecal continence. Moreover, the morbidity was similar with the two techniques. Then, new studies have confirmed the success rate of this device, in 23 patients with a median follow-up of 17.6 mo in the study of Barussaud et al[124] and in 18 patients with a follow-up from 353 d to 738 d in the study of Pakravan et al[125]. In both studies, symptoms, FI severity scores and QOL were significantly improved. However, although it brings promising results, the efficacy of the magnetic anal sphincter needs to be confirmed in larger and randomized studies with longer follow-up.

Ger et al[126] published the first report of the transposition of the antropyloric valve as a living sphincter at the end of an ileostomy in 1982. After this first publication, successive other works demonstrated the feasibility and the interest of this technique in animals[127-129]. More recently, some studies have been published about this technique in adult patients. The first preliminary report in human was published in 2011[130]. Then, Goldsmith et al[130] published successive studies on this technique demonstrating the feasibility and the success rate of this technique on clinical and manometric parameters in patients requiring anal replacement. In particular, they demonstrated in a study in 17 patients, after a median follow-up of 18 mo, a definite tone of the transposed graft on digital examination, an improvement in the St Mark incontinence score and in the QOL score (SF-36) associated with a significant rise in the postoperative resting neosphincter pressure[131-133].

However, although this new technique is very interesting, it remains a surgical invasive approach and it is, for the moment, at a very early stage of its development. Further larger studies with long-term follow-up will be necessary to evaluate the validity and the place of this technique.

Injection of Botulinum Toxin (BT) into the detrusor muscle is used for years by urologists to improve overactive bladder. In FI, especially in urge incontinence, a same mechanism may be hypothesized based on a rectal contractile disorder. A first study has been published to assess the efficacy of intrarectal injections of BT in the treatment of FI[134]. This prospective pilot study included 6 patients with high-amplitude contractions of the intact native rectum or of the reservoir (4 patients had a proctectomy for rectal cancer). Anal sphincters were intact in most of the patients. In this study, all the patients reported a clinical improvement based on the Cleveland Clinic Score at 3-6 mo that was sustained at 6 mo. Manometric data showed a decrease of the mean amplitude of contractions whereas the frequency of contractions remained unaffected by the BT injections. These results are interesting and encouraging, especially because it is a simple and non-invasive treatment. However, its efficacy needs to be demonstrated in larger studies with selected patients.

A vaginal bowel-control system has been reported in the treatment of FI. This device (Eclipse System) is a non-invasive, non-surgical therapeutic option. It consists of a vaginal insert with a pressure-regulated pump. The insert is made of a silicone-coated stainless steel base balloon that is posteriorly directed. In the prospective study of Richter including 110 patients, the intention-to-treat success rate at 1 mo was 78.7% and the success rate at 3 mo was 86.4%[135]. A significant improvement was also observed and no serious adverse effects were reported (mainly pelvic cramping or discomfort). This device is simple to use and non invasive and self-managed by the patients. Further studies will be necessary to determine the place of this device in the therapeutic algorithm of FI.

The effect of acupuncture on FI has been reported by an Italian study[136]. In this pilot study, 15 female patients were submitted to one acupuncture treatment per week for a 10-wk period, and a control session was repeated once per month up to 7 mo for six patients. After the 10-wk period, a significant improvement was observed with an overall mean continence score in the 15 patients from 10 (3-21) to zero (0-7). The continence index available in 14 patients at about 18 mo after start of treatment was 1 (0-8). Concerning manometric parameters, a significant increase was observed in the resting anal pressure and in the ability to sustain the squeeze pressure whereas the maximal sphincter squeeze pressure remained unchanged.