Comparison of antegrade continence enema treatment and sacral nerve stimulation for children with severe functional constipation and fecal incontinence

Abstract Background To compare antegrade continence enema (ACE) treatment and sacral nerve stimulation (SNS) in children with intractable functional constipation (FC) and fecal incontinence (FI). Methods We performed a retrospective review of children 6‐18 years old with FC and FI treated with either ACE or SNS at our institution. We recorded symptoms at baseline, 6 months, 12 months, 24 months, and their most recent visit after starting treatment. We compared improvement in FI, bowel movement (BM) frequency, abdominal pain, laxative use, and complications. Patients were contacted to evaluate perceived benefit using the Glasgow Children's Benefit Inventory. Key Results We included 23 patients treated with ACE (52% female, median age 10 years) and 19 patients treated with SNS (74% female, median age 10 years). Improvement in FI was greater with SNS than ACE at 12 months (92.9% vs 57.1%, P = .03) and 24 months (100% vs 57.1%, P = .02). Improvement in BM frequency was greater with ACE, and children were more likely to discontinue laxatives at all follow‐up time points (all P < .05). Improvement in abdominal pain was greater with ACE at the most recent visit (P < .05). Rate of complications requiring surgery was similar between groups (26.3% vs 21.7%). Benefit was reported in 83.3% and 100% of ACE and SNS groups, respectively (NS). Conclusions and Inferences Although both ACE and SNS can lead to durable improvement in children with FC and FI, SNS appears more effective for FI and ACE more effective in improving BM frequency and abdominal pain and in discontinuation of laxatives.


| INTRODUC TI ON
Functional constipation (FC) is a common condition in childhood with a described pooled prevalence of 9.5%. 1 The diagnosis of FC is made through fulfillment of the Rome criteria. 2 The majority of children with FC respond well to conventional medical and behavioral treatment strategies, including laxatives, dietary, and lifestyle changes. 3 However, a sizable number of children remain symptomatic despite optimal conventional treatment and are considered to have intractable FC. 3 Unfortunately, to date, treatment options for children with intrac-  The relatively novel SNS treatment involves electric stimulation of the sacral nerve root by an implanted lead connected to a pulse generator battery and is thought to modulate the function of the bowel, bladder, and/or pelvic floor. Sacral nerve stimulation has been shown to be effective in treating both urinary and fecal incontinence (FI) in adults. 6 However, the efficacy of SNS in the treatment of constipation is questionable 7,8 and experience with SNS in children remains very limited. A small number of studies have reported positive short-and long-term outcomes for children with constipation and FI treated with SNS, [9][10][11] but further research is needed.
The roles of ACE and SNS in the treatment of children with intractable FC therefore remain unclear, and guidelines for surgical treatment of intractable FC are lacking, leading to a wide variation in treatment practices between centers. 12 No studies have yet compared outcomes of ACE and SNS. The objective of this study was to compare the efficacy and safety of ACE and SNS treatment for children with intractable FC and to assess perceived health-related benefit and satisfaction.

| Study design and participants
This study consisted of two parts; first, we performed a retrospective cohort study comparing clinical symptoms and complications after starting ACE or SNS treatment. Next, we contacted all patients and parents included in the retrospective review to administer questionnaires assessing patient health-related benefit and satisfaction. Our study protocol was approved by the local institutional review board.
We included children between 6 and 18 years with clinically confirmed FC and FI based on the Rome III criteria who were treated with either ACE or SNS at Nationwide Children's Hospital in Columbus, Ohio, USA from 2012 through 2016. Children with organic causes of constipation or with prior abdominal surgery were excluded. For all patients, the decision to proceed with either ACE or SNS was made by the treating physician and family.

| Retrospective cohort study
Once we identified children meeting our inclusion and exclusion criteria, we recorded demographic information, medical and surgical history, and results of relevant diagnostic testing at baseline before ACE or SNS procedures. For each patient, information about clinical symptoms and complications was collected at 6 months, 12 months, 24 months, and at the most recent visit after starting ACE or SNS treatment. The most recent visit was defined as the latest follow-up visit in the medical chart at which the patient was still receiving ACE or SNS treatment.

Key Points
• Antegrade continence enemas (ACE) and sacral nerve stimulation (SNS) have both been described as treatment options for children with intractable constipation.
• In our retrospective study, SNS led to greater improvement in FI, but ACE was more effective in improving bowel movement frequency, and decreasing laxative usage.
• Treatment of children with intractable constipation and FI should be individualized based on presentation.
Larger, randomized studies are needed to better understand the roles of ACE and SNS.

| Antegrade continence enema procedure
Children undergoing ACE treatment underwent either a Malone appendicostomy procedure or a percutaneous cecostomy procedure. Malone appendicostomy procedures were performed by a pediatric surgeon and involved connection of the appendix to the abdominal wall to create a valve for catheterization and ACE administration. Percutaneous cecostomy procedures were performed by an interventional radiologist and involved the percutaneous introduction of a cecostomy tube into the cecum for ACE administration. Specific ACE flush components were determined by the treating physician for each individual patient and generally consisted of a combination of normal saline or polyethylene glycol solution and a stimulant laxative (glycerin or bisacodyl).

| Sacral nerve stimulation procedure
The SNS procedures were performed by either a pediatric surgeon or pediatric urologist and were all done in two stages. The first stage involved placement of a lead at the S3 sacral nerve root connected to a temporary pulse generator that remained external to the patient for a 2-week trial period. If clinical improvement was noted during this trial period, the patient underwent the second stage procedure.
The second stage involved the implantation of a permanent pulse generator battery (InterStim® System, Medtronic, Inc) into the subcutaneous fat of the upper buttock. The distribution and amplitude of nerve stimulation was determined by the treating physician in order to achieve an effective and comfortable stimulation.

| Clinical outcomes
We compared improvement in FI frequency, bowel movement frequency, abdominal pain, and oral/rectal laxative use at each followup time point for both treatment modalities. We defined successful treatment of FI as having FI less than once per week. We considered a bowel movement frequency of greater than twice per week as normal. 2 We defined successful treatment of abdominal pain as having pain less than once per week. Improvement in laxative usage was defined as complete discontinuation of oral and rectal laxatives.

| Patient benefit and satisfaction
To evaluate perceived patient benefit from each treatment, parents were contacted by telephone after the most recent follow-up time point and asked to complete the Glasgow Children's Benefit Inventory (GCBI). The GCBI is a validated measure of health-related benefit and contains 24 questions divided over four subscales (ie, "Emotion," "Vitality," "Learning," and "Physical health"). Items are scored on a five-point Likert scale ranging from "much worse" to "much better." Total and subscale scores are then transposed to a benefit scale ranging from −100 (maximum harm) to +100 (maximum benefit). A GCBI score >0 indicates positive health-related benefit. 12 In order to assess patient satisfaction, we asked parents two questions: (a) Whether they would proceed with the treatment again if given the chance to remake their decision and (b) whether they would recommend the treatment to other families.

| Statistical analysis
Chi-square and Fisher's exact tests were used to compare outcomes of ACE vs SNS treatment at each follow-up time point. Unpaired t tests and Mann-Whitney U tests were performed for the com-

| Baseline characteristics
All patients fulfilled the Rome III criteria for FC and had symptoms for >12 months (mean 66 months) and were treated with oral laxatives before ACE or SNS treatment. More patients treated with ACE had been previously treated with rectal enemas compared to patients treated with SNS (87.0% vs 5.3%, P < .01). Patients treated with SNS were more likely to have concomitant urinary symptoms compared to patients treated with ACE (94.7% vs 30.4%, P < .01).
Patient characteristics are shown in Table 1.

| Baseline diagnostic testing
Additional testing before the surgical procedure is displayed in Table 2.
Significantly more children in the ACE group underwent additional diagnostic testing as compared to the patients treated with SNS (100% vs 73.7%, P < .01). One child in the ACE group with a prior normal colonic transit time also had a normal bowel movement frequency at baseline.

| Bowel movement frequency
At baseline before starting ACE or SNS, 10

| Abdominal pain
As depicted in Table 3

| Complications
Overall complications were more common in the ACE group compared with the SNS group (  commented that their child's symptoms did not improve significantly. Another study reported a significant increase in bowel movement frequency after just 3 weeks of SNS treatment (5.9 vs 17.4, P < .01). 10 However, a long-term study at our institution showed no significant improvement in defecation frequency after more than 2 years of follow-up. 11 Randomized controlled trials in the adult population show similar negative results. 19,20 In contrast, in the ACE group, the number of children with a normal defecation frequency increased to 100% within the first 6 months. Long-term follow-up of patients after ACE showed similar high rates of improvement of symptoms up to 80%. 21 Our results therefore support the use of ACE treatment for children with intractable FC.

| D ISCUSS I ON
The differences between the effects of ACE and SNS in treating FC and FI are potentially secondary to different pathways of action.
Treatment with ACE works through mechanical irrigation of bowel contents, potentially in conjunction with stimulation of propagating colonic contractions, allowing the colon to fully evacuate on a regular basis and leads to improvement in FI by preventing stool accumulation. Although the precise mechanism by which SNS leads to improvement in FI remains incompletely understood, there is evidence that SNS modulates anorectal function, both centrally and peripherally. 22 Studies in adults showed that SNS may affect colonic motility by increasing the frequency of both antegrade and retrograde propagating pressure waves in patients with slow-transit constipation. 23,24 These effects on anorectal function or increased retrograde motor function in the colon may explain its ability to decrease FI. 22,25 The possibility remains that the described differences in clinical outcomes between ACE and SNS treatment may be in part second- with SNS treatment can be substantial. 31 We therefore stress that patients and their families should be educated about the possible risks of both treatments in order to make an informed decision.
Despite the associated complications and financial cost, families generally viewed ACE and SNS favorably. Although we were only able to contact half of our study population, we showed high patient perceived benefit and satisfaction after more than 4 years of treatment. More importantly, we found no significant differences in perceived benefit scores and satisfaction scores between children treated with ACE and SNS. These results support the use of both treatment strategies for children with severe FC and FI. However, owing to the small study sample and potential selection bias, prospective studies comparing quality of life, perceived benefit, and satisfaction after both surgeries are needed.
In conclusion, this retrospective comparison shows that both ACE and SNS treatments can be effective for children with intractable FC and FI. The ideal treatment option for each child should be based on his or her personal clinical symptoms. Our findings suggest that children with severe FI and concurrent urinary symptoms may benefit more from SNS treatment, while children who struggle primarily with stool evacuation and abdominal pain may benefit more from ACE treatment.
Obviously, both ACE and SNS should only be considered in patients with severe symptoms refractory to conventional treatment. Although considered minimally invasive, both therapies require surgical procedures with risk of severe complications. Prospective randomized studies comparing outcomes after ACE and SNS in a larger, homogenous cohort of children with FC are needed to better understand the optimal treatment strategy for children with intractable FC and FI.

D I SCLOS U R E
The authors have no financial relationships relevant to this article to disclose. The authors have no conflicts of interest to disclose.