FI is defined as a recurrent uncontrolled passage of fecal material over the prior 3 months in an individual with a developmental age of at least 4 years [1]. For research studies of this disorder, it is recommended that the onset of symptoms be considered for at least 6 months previously and confirmed cases should undergo 2−4 episodes of FI over any 4-week period.

Several community-based large-scale studies in the United States have suggested that FI is common, with a prevalence ranging from 7% to 15% in community-dwelling women, 18% to 33% in hospitals, and 50% to 70% in nursing homes [4,5]. The FI prevalence is generally found to be similar or lower in men than in women [2,6,7]. Interestingly, most FI patients seen in the clinic are women. In Korea, large-scale studies are rare. Among those that have been conducted, the incidence of incontinence in adult Korean was reported at 6.4% with a higher frequency in individuals over 50 years of age [8]. Another Korean study has described a FI prevalence of community-dwelling people aged 60 and older was 15.5% [9]. It should be noted that FI prevalence can be affected by population characteristics and statistical data, and by differences in definition, survey methods, screening questions, and differences in the reference time frame (e.g., 1 year or prior month) [2,10].

The etiology of FI is multifactorial [1,9]. It is thus more appropriate to focus on the relationship between the associated conditions and risk factors for this disorder, especially prior to its onset. In community surveys of bowel disturbances, particularly diarrhea and those involving rectal urgency, the burden of chronic disease was a more important and independent risk factor for FI than obstetric-related pelvic injuries (e.g., forceps, complicated episiotomy) [2,11,12]. An older age, watery diarrhea, and functional diarrhea have been identified as independent predictors in the Korean population [8]. Diabetes can also contribute to FI by weakening the sphincter or pelvic floor via anatomical defects, nerve damage or microvascular complications, and increases the risk of this disease in older women [13,14].

In a prior study of a community-based cohort, current smoking, body mass index, diarrhea, irritable bowel syndrome, cholecystectomy, and rectal prolapse were found to be independent risk factors for FI, with cholecystectomy and smoking observed to be more relevant than obesity [12]. Smoking, external sphincter atrophy, and obesity are also reported risk factors for FI [13,15]. Other conditions associated with FI include age, disease burden (number of accompanying events, diabetes), anal sphincter trauma (obstetric injury, prior surgery), and reduced physical activity [12-14]. Anorectal sensorimotor dysfunctions and/or altered bowel habits are also associated with FI in clinical practice (Table 1). Some of these conditions have not appeared to be a risk factor in community studies, which may be due to their relatively low prevalence. Consistent with the results of community-based studies, the majority of women with FI who consult physicians may not have a nervous system or inflammatory disorder but may have bowel disturbance disorders (usually diarrhea) associated with a history of obstetric risk factors. Notably however, neurological deficits can be identified only by neurophysiological tests which are not widely available. In a series of studies, the incidence of FI after vaginal delivery was less than 10%. There was no obvious difference between cesarean section and vaginal delivery. This may reflect a reduction in the use of vaginal delivery devices (e.g., forceps), a decrease in episiotomy procedures, and changes to cesarean delivery protocols [16,17]. The risk of FI is higher in women who undergo device-assisted deliveries, but whether or not the obstetric damage affects FI is not clear [12,15].

Anorectal surgery for a fistula, fissure, hemorrhoids, or resection of an anorectal carcinoma can damage the sphincter. In addition, impaired rectal compliance, as can occur following proctitis or after the creation of a pouch, and fecal impaction with diarrhea overflow can all cause FI [1].

A key symptom of FI is the loss of restraint in relation to various contents of the rectum i.e., solid stool, liquid/semi-formed stool, and gas. The outcomes of FI can be generally classified as staining, soilage, seepage, and accidents. Treatments may be changed due to pelvic floor disorders, rectal surgery, radiation therapy, hemorrhoid protrusion, and iatrogenic anal injury [29] but it is important also to distinguish between FI subtypes. FI without any awareness is referred to as passive FI, and leakage with self-awareness despite a positive measure is referred to as urge FI. In the former case, the FI correlates with sensory impairment, rectoanal reflex disorder, or abnormal function of the anal sphincter. The latter indicates that the sphincter has an abnormal function, or the rectum has a decreased capacity. Stool leaking after defecation may occur if the stool is incompletely discharged or rectal sensation is impaired.

The extent of symptoms in FI cases should be quantified using one of the validated scoring systems. Several measurement tools of proven efficacy for assessing the severity of symptoms and analyzing the effects of FI have been introduced [33,34]. However, this type of assessment is complicated by a striking absence of standardized definitions and of instruments that can be used to measure and quantitate FI. Although numerous scoring systems are commonly used (e.g., the Wexner/Cleveland Clinic Florida incontinence score, Fecal Incontinence Quality of Life questionnaire, Fecal Incontinence Severity Index, and St. Marks incontinence score) [33], none include physiologic components or objective test parameters that can accurately reflect the clinical severity of FI. Instead, most instruments are based on a subjective patient-reported assessment of severity and frequency.

Because FI can be caused by systemic or neurological diseases, both physical and neurological examinations should be performed.

If there is uncertainty in the diagnosis of FI after history taking and physical examination, and when several factors—such as overflow incontinence, decreased rectal capacity, and FI due to neurological causes —are excluded, further tests can be performed to analyze the structure and function of the anorectal lesion (Table 2).

Endoscopic evaluation of the rectosigmoid region is recommended to rule out colonic mucosal inflammation, rectal mass, or stricture, which can be achieved by performing flexible sigmoidoscopy. However, a colonoscopy may be more appropriate, particularly in an older individual. If patients have FI with diarrhea, endoscopic examination should be conducted to rule out inflammation of the mucous membranes, tumors, or narrowing of the rectum. Blood tests are then recommended to assess for thyroid function or other metabolic diseases as well as stool analysis.

Endoanal sonography is widely used as it is the most sensitive test for assessing sphincter defects or structural changes. Moreover, it can identify the thickness and structural deformities of the EAS and IAS [38] and is effective in detecting the rectal wall and puborectalis muscle. In particular, the endoanal sonography findings for an anal sphincter deformity have been found to be in good agreement with those from anorectal manometry [39]. EAS defects are detected with extremely high accuracy with this methodology. Sultan et al. [40] have used this technique to show that about 35% of parous women have defects in the EAS. If rectification of the sphincter injury is considered, sonography is particularly useful for identifying the structure of the sphincter. If sphincter injury is then found, evaluation of the EAS and electromyography of the puborectalis muscle can be helpful because they can rule out any accompanying denervation (a condition which decreases the surgical success rate in the experience of many clinicians).

Anorectal manometry and rectal sensory tests are effective in determining the function of the EAS/IAS and rectal sensory abnormalities. Prior studies have aimed to identify the rectal muscle strength as a reservoir through an objective understanding of the anorectal sensation, degree of endurance to stool volume, and rectal compliance [41-44]. Moreover, anal pressure measurement can be useful for biofeedback therapy [45,46]. In general, rest pressure, compression pressure, rectal sensation, rectal-anal reflex, and rectal compliance of the anal sphincter are measured (Fig. 1). Patients with FI have low anal sphincter pressure or decreased rectal sensations.

Anorectal manometry tests the magnitude of the resting anal pressure at a high-pressure zone (85% coming from rhythmic slow wave contractions of the IAS and 15% from tonic contraction of the EAS). The most common method is the use of a water-perfused catheter containing 4 recording channels to detect pressure at various points along the rectum/anal canal, with a balloon at the end. After testing baseline resting pressures, the patient is instructed to cough and then to squeeze the EAS, which provides the voluntary squeeze pressure. The pressures should increase quickly to prevent incontinence. The rectal balloon is then distended with fluid to elicit a quick decrease in anal pressure, indicating competency of the sampling reflex.

Electromyography is another method that can be used to detect nerve damage. Because denervation of a skeletal muscle is accompanied by re-innervation from neighboring axons, a single-fiber EMG electrode can detect multiple axons firing within a small area of the muscle to indicate whether it has been damaged and re-innervation has occurred.

Magnetic resonance imaging (MRI) of the anus or rectum has been recommended as an effective method for identifying the EAS, rectum, or structure around the rectum [47]. Among the types of MRI, dynamic MRI can provide good information about the anatomy and function of the pelvic floor but is not widely used as it is not cost-effective. Other possible tests may include changes in the anorectal angle, descent of the pelvic floor, length of the anal tube, rectocele, rectal prolapse, and rectal intussusception.

The initial treatment of FI is to educate the patient about the bowel function and the disease. By allowing patients to better understand FI, they can increase the effectiveness of their treatment and improve compliance. Although there is not yet enough evidence to show the effects of lift style modifications on FI, they are still recommended. Weight reduction also seems to be effective because obesity is a risk factor in FI as mentioned above, although no relationship between weight reduction and an FI decrease has been reported. A relative FI risk reduction of 20% (P=0.29) has been reported after bariatric surgery [48]. To prevent FI, it is also important to maintain regular bowel movements and defecation times and processed foods are not recommended. Patients are also recommended to keep a daily food diary to determine whether certain foods exacerbate FI symptoms. Bananas and potatoes are known to be helpful for loose stools and the intake of water within 30 minutes after meals can promote the movement of food in the intestines. However, this can promote food passage and worsen FI. It is thus recommended to wait for 30 minutes after eating before drinking. Many FI patients take laxatives due to constipation but this will prevent a constant stool. Augmented fluid intake and fiber-rich foods will typically improve constipation. A previous prospective study in older women reported that increased long-term dietary fiber intake decreased the risk of FI but other dietary treatments showed no significant effect [49].

As with urinary incontinence, Kegel exercises can help with FI. Pelvic floor muscle exercises after colo-anal anastomosis can improve FI symptoms and related QoL levels in colorectal cancer patients [50]. Kegel exercises can also prevent urinary and fecal incontinence in antenatal and postnatal women [51]. Bowel movement modification treatments include biofeedback treatment and a retraining program, which involves relearning by patients to assist with regular, effective and complete bowel movements. This can be done by inserting instruments or balloons with an electric sensor into the anus to strengthen the anal muscles. This procedure is used also to restore the sensation of the rectum [52]. The success rate of biofeedback treatments are reported to be 50%–90% [53]. However, if the pelvic floor is totally denervated or the capacity of the rectum is reduced due to surgery, the patient may be unresponsive.

Antimotility agents should be administered if intestinal motility is increased and incontinence occurs, and caffeine, alcohol, milk, and carbonated drinks should be avoided as much as possible. It is important to achieve a regular and predictable pattern of bowel movements. Complete emptying of the rectum is the best approach to preventing FI and this can be assisted by increasing dietary fiber intake, enemas or by stimulating the rectum [49]. Loperamide may be effective in diarrhea-related FI patients [54]. Small-scale trials, have also found that loperamide decreases FI symptoms. Loperamide and psyllium have both been found to reduce FI symptom severity but the loperamide group experienced more constipation (29% vs. 10%) [55].