For the patient with moderate to severe fecal incontinence and a demonstrated sphincter defect in whom best conservative management has not produced sufficient improvement, anal sphincter repair may be considered. A number of authors have reported good to excellent short term results in 60–80 % of patients as evaluated by follow-up questionnaires and QOL measures over 35 years of accumulated data (See Table 39.1) [9–18]. However, the reports are generally small series with retrospective data collection, and few have any comparison groups. It is important to note that most surgeons exclude gaps over 120° from repair.

Important information regarding the longer term durability of anal sphincter repair has been accumulated in the past 13 years [17–23]. The proportion of patients reporting good to excellent outcomes in the long term, approximately 10 years after sphincteroplasty, varies from 14 to 62 % (See Table 39.2.). This is significantly less than that reported in the short term, suggesting that function after anal sphincter repair tends to deteriorate over time. However, available literature is limited, and as with the data on short term outcomes, these reports are mostly small series with retrospective data collection and few have any comparison groups. In addition, patient populations and methods of assessing outcomes are heterogeneous, which makes it difficult to draw clear conclusions.

With the realization that long term outcome after anal sphincter repair appears to deteriorate over time, a number of authors have attempted to define what variables might predict outcome. Of nine studies that evaluated the effect of age at surgery on outcome [17, 20–22, 24–27], four [20, 22, 24, 27] reported poorer outcome in patients over age 50. One [24] of three studies that looked at parity [17, 24, 28] found that patients with a history of two or more vaginal births had a poorer outcome after sphincteroplasty. Obstetric injury etiology was associated with better outcome compared to other causes of incontinence in one study that commented on this [22]; however, the number of patients with differing etiologies was small. An observational study of estrogen therapy in postmenopausal women with fecal incontinence found symptomatic improvement in 90 % after 6 months of hormone replacement therapy, with increases in resting and squeeze pressures and an increase in maximum tolerated rectal volume [29]. The patients with an identifiable sphincter defect had no difference in outcome. However, the potential application of this to the population of patients who are candidates for sphincteroplasty is not clear. All authors felt that older patients should still be considered for anal sphincter repair, though the risk of a possible inferior outcome should be discussed.

Several studies assessed outcome related to initial physiologic and anatomic variables, including resting and squeeze pressures, anal canal length, rectal compliance, pudendal neuropathy, and presence of internal anal sphincter defect [17, 19–21, 25, 28]. None found resting or squeeze pressures, anal canal length, rectal compliance, or presence of internal anal sphincter defect predictive of outcome. Only one [30] of the five studies that evaluated pudendal neuropathy [17, 21, 25, 28, 30] reported that this was predictive of a poorer outcome. However, all authors felt that anal sphincter repair should be offered to patients with pudendal neuropathy, with a discussion of possible poorer outcome.

There is little solid information regarding predictive value of technical aspects of anal sphincter repair. Most surgeons perform overlapping repair. One study [22] evaluated outcome after overlapping repair vs end-to-end repair of the sphincter. No predictive effect was identified. Maslekar et al. [21] felt their good results (86 % good outcome at 7 years) were related to their technique of dissecting each sphincter separately, though they did not include any comparison group. Three studies compared outcomes using fecal diversion with repair performed without diversion [28, 30, 31]. No predictive effect regarding outcome was seen; fecal diversion is not routinely used in anal sphincter repair in the United States.

There does appear to be a predictive relationship between short and long term outcomes. Vaizey et al. [31] found that patients who had good outcomes in short term tended to have more durable outcomes, compared to those who did poorly initially. Malouf et al. [28] reported that the Parks score at 15 months after anal sphincter repair was predictive of long term success. Bravo Gutierrez et al. [20] found that a poor outcome at 3 years after surgery was a strong predictor of poor long term outcome.

Identification of a recurrent or persistent anal sphincter defect after sphincteroplasty is important, as these patients may be offered a repeat anal sphincter repair. Giordano et al. [32] reported on 36 patients who underwent a repeat sphincter repair after demonstration of a persistent sphincter defect. The repeat repair group reported good (50 %) and adequate (11 %) function at a median of 20 months, compared with the patients undergoing first-time repair (58 % good, 17 % adequate). Vaizey et al. [31] reported on 23 patients with a repeat anal sphincter repair. Twenty-one were evaluable at 20 months after repeat repair. One was fully continent, 12 reported 50 % or more symptom improvement over preoperative function, and four were unchanged. Hong et al. [33] reported retrospectively on 59 patients with failed sphincteroplasty. In this cohort, 33 underwent repeat sphincteroplasty, 11 had artificial bowel sphincter (ABS) implant, and 15 underwent sacral nerve stimulation (SNS). Observed improvements in continence were similar; however, the rate of complications and reoperations was significantly lower in the repeat sphincteroplasty group, leading the authors to suggest that repeat sphincteroplasty should be considered the first choice in the management of failed anal sphincter repair. However, function after repeat sphincteroplasty may deteriorate over time more markedly than after a first repair [34].

Comparing outcomes of anal sphincter repair is made more difficult by the heterogeneity of the measures utilized in reporting. Many methods of assessment are common, and while those most widely used contain elements of incontinence frequency and severity, not all include patient-defined quality of life measures. However, the quality of life determination may be the most important consideration for the patient who must weigh the effect on daily life activities against the possible risks of treatment. Evaluation of function in 62 patients a mean of 70 months after anal sphincter repair showed that while 70 % had objective clinical improvement, only 55 % considered their bowel control improved and only 45 % were satisfied with the outcome [17]. The authors note that urgency was the most important symptom related to patient satisfaction after anal sphincter repair: 24 of 26 patients in whom urgency had improved reported that they were happy with the outcome.