- Email: [email protected]
Sacral Neuromodulation in Children With Urinary and Fecal Incontinence: A Multicenter, Open Label, Randomized, Crossover Study
Sacral Neuromodulation in Children With Urinary and Fecal Incontinence: A Multicenter, Open Label, Randomized, Crossover Study M. Haddad,* R. Besson, D. Aubert, P. Ravasse, J. Lemelle, A. El Ghoneimi, J. Moscovici, F. Hameury, K. Baumstarck-Barrau, G. Hery and J. M. Guys From the Departments of Pediatric Surgery, Hôpital Jeanne de Flandre, Lille (RB), Hôpital Saint Jacques, Besançon (DA), CHU Caen, Caen (PR), CHU Nancy, Nancy (JL), Hôpital Robert Debré, Paris (AEG), Hôpital des Enfants, Toulouse (JM), CHU Lyon, Lyon (FH), and Hôpital d’Enfants de la Timone (MH, GH, JMG) and Department of Public Health, Université de la Méditérranée (KB-B), Marseille, France
Abbreviations and Acronyms CIC ⫽ clean intermittent catheterization SNM ⫽ sacral neuromodulation Submitted for publication October 30, 2009. Study protocol was approved by a French ethics committee and registered as a clinical trial (AFSSAPS 2005/04/003) at the French Ministry of Health. Study was supported by grants from French Ministry of Health (PHRC 2004 program hospitalier de recherche clinique). * Correspondence: Department of Pediatric Surgery, Hôpital d’Enfants de la Timone 264, rue Saint Pierre, 13005 Marseille Cedex 5, France (telephone: 33-4-91-38-66-82; FAX: 33-4-91-3847-14; e-mail: [email protected]).
See Editorial on page 417.
Purpose: The clinical benefit of sacral neuromodulation is unclear due to the paucity of randomized trial data. The purpose of this study was to evaluate sacral neuromodulation for management of urinary and fecal incontinence in a pediatric population. Materials and Methods: This multicenter, open label, randomized, crossover study included children older than 5 years. After trial stimulation of the S3 root a neuromodulator (InterStim®) was implanted on the S3 foramen. Clinical examinations, voiding and bowel diaries, and urodynamic and manometric evaluations were performed at the beginning (t1) and end (t2) of the first period, and at the beginning (t3) and end (t4) of the second period. Results: A total of 33 patients (24 boys) with a mean ⫾ SD age of 12.22 ⫾ 5.09 years were randomized. Etiologies were mainly of neurological origin. Incontinence was mixed urinary and fecal in 19 cases, urinary only in 9 and fecal only in 5. Cystometric bladder capacity increased during sacral neuromodulation (delta ⫹24.27 ml vs ⫺37.45 ml, p ⫽ 0.01). There was no significant change in other urodynamic or manometric parameters. Overall positive response rate was more than 75% for urinary (81%) and bowel (78%) function. Crossover analysis indicated that sacral neuromodulation is more effective than conservative treatment for both types of incontinence (p ⫽ 0.001). Conclusions: In a pediatric population sacral neuromodulation is effective for bladder and bowel dysfunction and should be considered before irreversible surgery. Key Words: congenital abnormalities, electric stimulation therapy, fecal incontinence, lumbosacral plexus, urinary incontinence
696
www.jurology.com
SPHINCTER disorders can have many etiologies in children but the most frequent (95%) is congenital vertebral and medullary malformation, eg spinal dysraphism.1 Neurogenic bladder is characterized by loss of ability to control voiding. With age most patients have obstructive uropathy with increase in bladder pressure during filling and vesicosphincteric dyssyn-
ergia. Since the lower urinary and gastrointestinal tracts have the same embryological origin and innervations, sphincter disorders are frequently associated with fecal incontinence (80%).1,2 SNM has been used for management of sphincter disorders in adults.3–8 However, only 3 groups have reported results in children.4,9 –11 Most pediatric
0022-5347/10/1842-0696/0 THE JOURNAL OF UROLOGY® © 2010 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION
Vol. 184, 696-701, August 2010 Printed in U.S.A. DOI:10.1016/j.juro.2010.03.054
AND
RESEARCH, INC.
SACRAL NEUROMODULATION, AND URINARY AND FECAL INCONTINENCE
studies have involved small patient populations and data have been mainly clinical. In a previous study describing SNM in children with neurogenic bladder, we reported improvement in bladder and bowel function.9 These encouraging results led to the present study, which was designed to evaluate the efficacy and tolerance of SNM for management of urinary and fecal incontinence in a pediatric population.
METHODS Patients and Setting This prospective, national, randomized, open-label, crossover study was conducted in the pediatric surgery departments of French university hospital centers in Marseille, Lille, Besançon, Caen, Nancy, Paris, Lyon and Toulouse. Patients with urinary incontinence due to neurogenic bladder and/or fecal incontinence due to congenital malformation were eligible. Children or teenagers under surveillance for fecal and/or urinary incontinence were included if they presented with at least 2 of the following conditions— duration of continence 90 minutes or less, post-void residual volume greater than 50% of functional bladder capacity, bladder compliance less than 15 and bladder overactivity with pressure peaks exceeding 40 cm H2O. Ability and motivation of patient and family to comply with keeping a voiding and bowel diary, and attend followup examinations throughout the study were also taken into account for inclusion. All treatment, especially anticholinergic medication, was discontinued during the screening period. The main exclusion criteria were local
697
risk factors for device implantation (scar and sacral agenesis involving greater than 50% of the sacrum) and failure to detect the S3 roots. Before inclusion all patients underwent renal ultrasound, urethrocystography and renal scintigraphy (dimercapto-succinic acid) to evaluate bladder and kidney status. Routine preanesthetic laboratory tests (including creatininemia) were performed preoperatively. If clinical status of the patient changed during followup, testing was repeated to detect deterioration. Written informed consent was obtained from the parents and patients according to age. The study protocol was approved by a French ethics committee and registered as a clinical trial (AFSSAPS 2005/04/003) at the French Ministry of Health.
Implantation Technique As described previously, S3 roots were stimulated percutaneously with the patients under general anesthesia.9 If responses were acceptable, the neurostimulation device and lead were implanted immediately. The test stimulation period used in adults was not applied because we consider it too short to induce a response in children with neurological conditions. Patients were discharged home after 48 hours, provided that general status and pain level allowed.
Randomization A randomization list was generated by computer using a permuted block design. Subjects were randomly assigned to 1 of 2 neuromodulation groups, 6 months ON followed by 6 months OFF (group A) or the opposite sequence (group B). The 2 phases were separated by a 45-day washout period to return to baseline status (see figure).
Stimulation of S3 root
Positive Implantation Lead in S3 foramen, generator S/C
Negative excluded
Randomisation
Group A (Stimulation: ON)
Group B (Conservative treatment: OFF)
End therapeutic phase 1: 7th month “Cross Over” Beginning therapeutic phase 2: 9th month Group A (Conservative treatment: OFF) End therapeutic phase 2: 15th month Study design
Group B (Stimulation: ON)
698
SACRAL NEUROMODULATION, AND URINARY AND FECAL INCONTINENCE
Stimulation Neurostimulation was activated 24 hours after implantation in group A or after the OFF phase in group B. The initial pulse duration setting was 210 milliseconds in the cyclic mode (ON/OFF, 28/8 seconds). At each followup pulse amplitude (range 1 to 9 V) and frequency (10 to 20 Hz) were adjusted to maintain desired motor responses.
Data Collection At enrollment in the study (baseline evaluation, t1) demographic, clinical and severity data were collected for each patient. Following implantation response was evaluated at the end of the first 6-month period (t2), and at the beginning (t3) and end (t4) of the second period. At each evaluation physical examination was performed, including assessment of clinical status based on voiding and bowel diaries, and measurement of urodynamic and rectomanometric parameters.
End Points Patients were classified as responders or nonresponders regarding urinary and fecal performance after each phase of the ON and OFF sequence. Response was defined as resolution of urinary leakage and/or fecal soiling with no need for pads, or a decrease of more than 50% in the number of leaks and/or soilings with minimum protection needed between the beginning and end of each sequence. All other findings were scored as “no response.” The primary end point was defined by the overall response determined by combining clinical urinary and fecal responses. SNM was considered effective if either urinary or fecal response occurred and ineffective if neither urinary nor fecal response occurred. Secondary end points were urodynamic and rectomanometric parameters. In all patients the evaluation technique consisted of insertion of a triple lumen 8Fr to 10Fr Bohler urethral catheter, slow infusion of saline (20 ml per minute) and recording of abdominal pressure. Urodynamic assessment was performed in accordance with International Continence Society guidelines and included measurement of leak point pressure, cystometric bladder capacity, bladder activity, sphincter pressure, bladder compliance and maximal cystometric bladder capacity.12,13 Fecal assessment included the parameters transit time, rectal volume, resting pressure and voluntary contraction pressure. Changes in each parameter were compared between the end and beginning of the period (t2 vs t1 or t4 vs t3).
Statistical Analysis The sample size was designed to obtain 80% power to detect a 40% difference in clinical response at 6 months between the 2 groups, considering a potential loss to followup rate of 20% and S3 root detection failure of 20%. Comparison of gender, age and type of incontinence between groups A and B was performed using Student’s t test or Mann-Whitney test for quantitative variables and chi-square or Fisher’s exact test for frequencies. Primary and secondary outcomes were compared between the ON and OFF periods. The primary outcome (clinical response) was analyzed using the McNemar test. Secondary outcomes (urodynamic and fecal parameters) were analyzed using the Wilcoxon signed rank test. The second analyses included 2 subgroups, ie urinary incontinence (isolated
urinary or mixed incontinence) and fecal incontinence (isolated fecal or mixed incontinence). Statistical analysis was performed using SPSS®, version 15.0 software. Differences were considered significant at p ⫽ 0.05.
RESULTS Sample A total of 41 patients underwent trial assessment between April 2004 and September 2007. The S3 root was detected in only 33 patients who were randomized (overall implantation success was 81%). There were 24 boys and 9 girls with a mean ⫾ SD age of 12.22 ⫾ 5.09 years. Incontinence was urinary only in 9 patients, fecal only in 5 and mixed in 19. A total of 17 patients with urinary incontinence were on CIC. The underlying etiologies were spina bifida in 10 patients, sacral agenesis in 8, miscellaneous neurological anomalies in 7 (including 2 tumors), and congenital coloanal and urinary malformations in 5. Patients were randomly divided into 2 treatment groups, with 16 in group A and 17 in group B. Three patients randomized to group B were treated using the group A protocol due to a mix-up in instructions given to 1 team. The final population was 19 patients in group A and 14 in group B. At inclusion the 2 groups were comparable regarding gender, age and type of incontinence (table 1). A total of 24 patients completed the ON and OFF periods. Six patients dropped out due to device infection in 3, degradation of urinary incontinence following deactivation that required reactivation in 1, device removal and refusal to continue the trial in 1, and severe complications of the initial neurological pathology (Wolfram syndrome) in 1. In the remaining 3 patients clinical response was not available at the end of either the first or the second period. Efficacy of SNM Clinical response was significantly better when SNM was ON than OFF (75% vs 21%, p ⫽ 0.001). No patient was scored as a responder when SNM was OFF and nonresponder when SNM was ON. Respondership was higher for urinary and fecal incontinence when SNM was ON than OFF (81% vs 24%, p ⫽ 0.001 and 78% vs 17%, p ⫽ 0.001, respectively). In the crossover model using the McNemar test comparison of clinical responses showed “stimulation Table 1. Characteristics of randomized groups
M-to-F ratio Mean ⫾ SD age (yrs) No. incontinence: Fecal Mixed Urinary
Group A
Group B
Total
p Value
5:14 12.21 ⫾ 4.72
4:10 12.23 ⫾ 5.73
9:24 12.22 ⫾ 5.09
1.00 0.99 0.88
3 10 6
2 9 3
5 19 9
SACRAL NEUROMODULATION, AND URINARY AND FECAL INCONTINENCE
Table 2. Global clinical response between ON and OFF treatments OFF
No. ON Response (%)
Response No response Total
No. ON No Response (%)
Total No. (%)
5 13
0 6
5 (21) 19 (79)
18 (75)
6 (25)
24 (100)
⫽ response and control ⫽ no response” in 13 patients and “stimulation ⫽ no response and control ⫽ response” in none (p ⫽ 0.001, tables 2 and 3). The crossover model was also used to compare changes in urodynamic and manometric variables through time during each therapeutic phase. This analysis took into account the impact of period and sequence on therapeutic benefit. A significant increase in cystometric bladder capacity was observed during stimulation (delta ⫹24.27 ml vs ⫺37.45 ml, p ⫽ 0.01). The bladder was significantly more overactive with than without neuromodulation (⫹1 vs ⫺0.36, p ⬍0.001). No significant difference was noted between other urodynamic and rectomanometric variables (table 4). Tolerance of SNM The procedure was well tolerated. No patients dropped out of the study due to worsening urodynamic parameters with upper tract deterioration. Infection was observed in 4 patients (all nonresponders) and required removal after 1 month in 1, after the first ON phase in 2 and at the end of the trial in 1. Lead revision was required in 2 responders. Overall complication rate was 18% (6 of 33 patients).
DISCUSSION This study demonstrates that SNM can improve urinary and fecal continence in children with congenital malformations. SNM was introduced in 1981 for treatment of stress urinary incontinence or dysuria in adults. A benefit for anal incontinence was discovered almost by chance in adults treated for urinary disorders. Following this discovery SNM has also been proposed for isolated anal incontinence since 1995.3 In 2004 we reported subjective improvement in intestinal transit in children with neurological conditions.9 In a subsequent study aimed at assessing SNM for dysfunctional elimination syndrome in children without neurological conditions Roth et al reported improvement in urinary incontinence and constipation in 88% and 71% of patients, respectively.10 The present study revealed almost identical or better results for fecal incontinence (78%) but slightly lower results for urinary incontinence (81%). The latter finding is probably due to the neurological origin of urinary incontinence, which
699
Scheepens14 and Keppene15 et al consider an unfavorable predictor, especially for extensive lesions with long-standing symptomatology (eg congenital malformations). Lombardi et al obtained good results in partial spinal cord injury when the time between diagnosis and SNM was short.16 This finding suggests that early implantation may maximize benefits. The clinical improvement in urinary incontinence observed in this study was not always correlated with improvement in urodynamic parameters. Others have reported the same disconnect for neuromodulation,4 as well as other therapeutic treatments such as bulking agents in the bladder neck.17,18 Nevertheless, our crossover analysis comparing urodynamic variables after the 2 phases of treatment showed a statistically significant favorable change in cystometric bladder capacity under SNM. Our findings also demonstrated some bladder overactivity during SNM compared to conventional therapy, probably in relation to discontinuation of anticholinergic therapy. This result was not related to clinical outcome. Improvement in fecal incontinence observed in this study similarly was not always linked to manometric findings. Previous reports have been contradictory in this regard. Several authors have noted no manometric improvement in adults treated with SNM for fecal incontinence related or unrelated to neurological malformations,16,19,20 while others have reported improvement. 21–23 In this series intestinal transit time was prolonged with SNM but the change was not statistically significant. This result is in accordance with the findings of Michelsen et al, who reported that SNM may prolong colonic transit time by reducing the antegrade propagating sequence frequency.24 The physiological mechanisms underlying the mode of action of SNM on fecal and urinary incontinence are still barely understood. Clinical and experimental studies have revealed that SNM impacts bowel frequency and rectal sensation but has only low effect on the compliance of the anal sphincter.20,25–27 Boyle et al showed that the outcome of SNM for treatment of fecal Table 3. Clinical response between ON and OFF treatments in urinary and fecal incontinence subgroups OFF
No. ON Response (%)
Isolated or mixed urinary incontinence: Response 5 No response 12 Total 17 (81) Isolated or mixed fecal incontinence: Response 3 No response 11 Total p ⫽ 0.001 (McNemar).
14 (78)
No. ON No Response (%)
Total No. (%)
0 4
5 (24) 16 (76)
4 (19)
21 (100)
0 4
3 (17) 15 (83)
4 (22)
18 (100)
700
SACRAL NEUROMODULATION, AND URINARY AND FECAL INCONTINENCE
Table 4. Comparisons between ON and OFF stimulations for urinary and fecal parameters ON (mean ⫾ SD delta)* Urinary parameters: Leak point pressure Cystometric bladder capacity Overactivity Sphincter pressure Bladder compliance Total bladder capacity Fecal parameters: Transit time Rectal vol Resting pressure Voluntary contraction pressure
OFF (mean ⫾ SD delta)*
Order Effect
Time Effect
Treatment Effect
7.33 ⫾ 20.90 24.27 ⫾ 84.18 1.00 ⫾ 17.75 7.35 ⫾ 29.46 1.04 ⫾ 12.54 23.38 ⫾ 98.71
⫺5.42 ⫾ 32.64 ⫺37.45 ⫾ 77.85 ⫺0.36 ⫾ 24.82 ⫺5.50 ⫾ 46.36 ⫺5.30 ⫾ 16.56 ⫺20.04 ⫾ 94.30
0.74 0.04 0.08 0.28 0.59 0.02
0.73 0.59 0.48 0.59 0.39 0.81
0.31 0.01 ⬍0.001 0.52 0.31 0.14
17.38 ⫾ 40.28 ⫺15.00 ⫾ 29.58 8.37 ⫾ 25.11 6.35 ⫾ 28.81
⫺24.13 ⫾ 76.22 ⫺4.72 ⫾ 49.22 ⫺3.16 ⫾ 28.31 ⫺3.26 ⫾ 26.94
0.56 0.20 0.86 0.95
0.68 0.77 0.75 0.18
0.10 0.63 0.20 0.46
* Difference in values at beginning and end of each therapeutic period.
incontinence is not affected by the presence or severity of anal sphincter defects.28 Other research projects using functional and standard magnetic resonance imaging focus on the effects of SNM in patients with overactive bladder or urinary retention. Blok et al reported that in patients with urge urinary incontinence SNM seems to modulate areas in the brain normally involved in alertness and awareness, ie having a role in storage and voiding.29 In a recently published study Kavia et al concluded that in patients with urinary retention and Fowler syndrome SNM inhibits the urethral afferents to enable voiding.30 All of our patients undergoing CIC, even responders, still required CIC despite SNM. In most adult studies deactivating the neuromodulator led to reappearance of symptoms, at least initially. However, in a study of 20 children with elimination disorders Roth et al observed 2 patients who remained asymptomatic after deactivation.10 In our series of 33 patients 5 reported continued response after deactivation (table 1). These findings suggest partial neurological deficit and a possible role of SNM in training faulty neural pathways responsible for the insult. One of our patients presented with bladder overactivity following sacral agenesis. This patient noticed improvement without urodynamic correlation and wished to remain under SNM at the end of study. Urodynamic findings in this patient normalized 1 year after the study. This result suggests that clinical response occurs earlier than urodynamic re-
sponse. A study with longer followup is planned to address these issues. Use of SNM in growing children poses the problem of lead migration and battery longevity. The lead design offers the possibility of 4 different stimulation points, which is compatible with some growth. Battery replacement is simple. The complication rate in this study was 18.8%. Two types of complications occurred, ie infection (4 cases) and electrode migration (2). Surgical intervention to remove the device or reposition the electrode was required in all cases. Comparable complication rates have been observed in previous pediatric reports with no deleterious long-term consequences.10 There was no evidence that complications were more common according to type of incontinence. Device miniaturization and lead redesign for children should reduce future technical problems and decrease morbidity.
CONCLUSIONS Numerous studies have confirmed the efficacy of SNM in urinary and digestive disorders in adults. The present study demonstrates that SNM can also be effective for neurogenic incontinence in children. Based on these findings, we recommend attempting SNM before irreversible surgery.
ACKNOWLEDGMENTS Prof. P. Auquier and Mr. A. D. Loundou, Department of Public Health, Marseille Medical School, assisted in statistical analysis.
REFERENCES 1. Nijman RJ and Tekgul S: Pathophysiology of neurogenic bladder dysfunction. In: Pediatric Neurogenic Bladder Dysfunction—Diagnosis, Treatment, Long-Term Follow-Up. Edited by C Esposito, JM Guys, D Gough et al. New York: Springer 2006; chapt 4, p 33.
2. Ponticelli A, Iacobelli BD and Silveri M: Colorectal dysfunction and faecal incontinence in children with spina bifida. Br J Urol, suppl., 1998; 81: 117. 3. Matzel KE, Stadelmaier U and Hohenfellner M: Electrical stimulation of sacral nerves for treat-
ment of faecal incontinence. Lancet 1995; 346: 1124. 4. Bosch JL and Groen J: Sacral nerve modulation in the treatment of patients with refractory motor urge incontinence: long-terms results of a pro-
SACRAL NEUROMODULATION, AND URINARY AND FECAL INCONTINENCE
spective longitudinal study. J Urol 2000; 163: 1219. 5. Indar A, Young-Fadok T and Cornella J: A dual benefit of sacral neuromodulation. Surg Innov 2008; 15: 219. 6. Tjandra JJ, Lim JF and Matzel K: Sacral nerve stimulation: an emerging treatment for fecal incontinence. Aust N Z J Surg 2004; 74: 1098. 7. Fandel T and Tanagho EA: Neuromodulation in voiding dysfunction: a historical overview of neurostimulation and its application. Urol Clin North Am 2005; 32: 1. 8. Bernstein AJ and Peters KM: Expanding indications for neuromodulation. Urol Clin North Am 2005; 32: 59. 9. Guys JM, Haddad M, Planche D et al: Sacral neuromodulation for neurogenic bladder dysfunction in children. J Urol 2004; 172: 1673. 10. Roth TJ, Vandersteen DV, Hollatz P et al: Sacral neuromodulation for dysfunctional elimination syndrome: a single center experience with 20 children. J Urol 2008; 180: 306. 11. Hoebeke P, Van Laecke E, Everaert K et al: Transcutaneous neuromodulation for the urge syndrome in children: a pilot study. J Urol 2006; 166: 2416.
701
14. Scheepens WA, Jongen MM, Nieman FH et al: Predictive factors for sacral neuromodulation in chronic lower urinary tract dysfunction. Urology 2002; 60: 598.
22. Matzel KE, Stadelmaier U, Hohenfellner M et al: Chronic sacral spinal nerve stimulation for fecal incontinence: long-term results with foramen and cuff electrodes. Dis Colon Rectum 2001; 44: 59.
15. Keppene V, Mozer P, Chartier-Kastler E et al: Neuromodulation in the management of neurogenic lower urinary tract dysfunction. Prog Urol 2007; 17: 609.
23. Ripetti V, Caputo D, Ausania F et al: Sacral nerve neuromodulation improves physical, psychological and social quality of life in patients with fecal incontinence. Tech Coloproctol 2002; 6: 147.
16. Lombardi G, Del Popolo G, Cecconi F et al: Clinical outcome of sacral neuromodulation in incomplete spinal cord-injured patients suffering from neurogenic bowel dysfunctions. Spinal Cord 2009; 11: 101. 17. Guys JM, Breaud J, Hery G et al: Endoscopic injection with polydimethylsiloxane for the treatment of pediatric urinary incontinence in the neurogenic bladder: long-term results. J Urol 2006; 175: 1106. 18. Lottmann HB, Margaryan M, Bernuy M et al: The effect of endoscopic injections of dextranomer based implants on continence and bladder capacity: a prospective study of 31 patients. J Urol 2002; 168: 1863. 19. Faucheron JL, Bost R, Duffournet V et al: Sacral neuromodulation in the treatment of severe anal incontinence. Forty consecutive cases treated in one institution. Gastroenterol Clin Biol 2006; 30: 669.
12. Abrahams P, Blaivas JG, Stanton SL et al: The standardisation of terminology of lower urinary tract function. Scand J Urol Nephrol 1998; 114: 5.
20. Uludag O, Koch SM, Van Gemert WG et al: Sacral neuromodulation in patients with fecal incontinence: a single-center study. Dis Colon Rectum 2004; 47: 1350.
13. Everaert K, Plancke H, Lefevere F et al: The urodynamic evaluation of neuromodulation in patients with voiding dysfunction. Br J Urol 1997; 79: 702.
21. Leroi AM, Michot F, Grise P et al: Effect of sacral nerve stimulation in patients with fecal and urinary incontinence. Dis Colon Rectum 2001; 44: 779.
24. Michelsen HB, Christensen P, Krogh K et al: Sacral nerve stimulation for faecal incontinence alters colorectal transport. Br J Surg 2008; 95: 779. 25. Dinning PG, Fuentealba SE, Kennedy ML et al: Sacral nerve stimulation induces pan-colonic propagating pressure waves and increases defecation frequency in patients with slow-transit constipation. Colorectal Dis 2007; 9: 123. 26. Uludag O, Morren GL, Dejong CH et al: Effect of sacral neuromodulation on the rectum. Br J Surg 2005; 92: 1017. 27. Vitton V, Gigout J, Grimaud JC et al: Sacral nerve stimulation can improve incontinence in Crohn’s patients with internal and external sphincter disruption. Dis Colon Rectum 2008; 51: 924. 28. Boyle DJ, Knowles CH, Lunniss PJ et al: Efficacy of sacral nerve stimulation for fecal incontinence in patients with anal sphincter defects. Dis Colon Rectum 2009; 52: 1234. 29. Blok BF, Groen J, Bosch JL et al: Different brain effects during chronic and acute sacral neuromodulation in urge incontinent patients with implanted neurostimulators. BJU Int 2006; 98: 1238. 30. Kavia R, Dasqupta R, Critchley H et al: A functional magnetic resonance imaging study of the effect of sacral neuromodulation on brain responses in women with Fowler’s syndrome. BJU Int 2010; 105: 366.
EDITORIAL COMMENT The authors are to be congratulated for this original study. One must remember that few studies are designed for the pediatric population, whatever the topic. The authors had the novel idea of focusing on global pelvic floor dysfunction (urinary and fecal incontinence), which increases the interest of the study. In a time of evidence-based medicine this is probably the only controlled and well shaped study to evaluate the effect of SNM on a neurogenic pediatric population. Despite the fact that SNM may or may not be effective at improving quality of life and urinary incontinence in an adult neurogenic popula-
tion,1 it is well-known that it can demonstrate individual usefulness. SNM is a conservative therapy that may act on urinary and fecal incontinence. New treatments are emerging for neurogenic urinary incontinence,2 provided that patients are using clean intermittent catheterization to empty the bladder. This series gives us a more clear and independent evaluation of SNM in a unique study design. Emmanuel Chartier-Kastler Department of Urology Pierre and Marie Curie Faculty of Medicine Paris, France
REFERENCES 1. Lombardi G and Del Popolo G: Clinical outcome of sacral neuromodulation in incomplete spinal cord injured patients suffering from neurogenic lower urinary tract symptoms. Spinal Cord 2009; 47: 486. 2. Game X, Mouracade P, Chartier-Kastler E et al: Botulinum toxin-A (Botox) intradetrusor injections in children with neurogenic detrusor overactivity/neurogenic overactive bladder: a systematic literature review. J Pediatr Urol 2009; 5: 156.
Abbreviations and Acronyms CIC ⫽ clean intermittent catheterization SNM ⫽ sacral neuromodulation Submitted for publication October 30, 2009. Study protocol was approved by a French ethics committee and registered as a clinical trial (AFSSAPS 2005/04/003) at the French Ministry of Health. Study was supported by grants from French Ministry of Health (PHRC 2004 program hospitalier de recherche clinique). * Correspondence: Department of Pediatric Surgery, Hôpital d’Enfants de la Timone 264, rue Saint Pierre, 13005 Marseille Cedex 5, France (telephone: 33-4-91-38-66-82; FAX: 33-4-91-3847-14; e-mail: [email protected]).
See Editorial on page 417.
Purpose: The clinical benefit of sacral neuromodulation is unclear due to the paucity of randomized trial data. The purpose of this study was to evaluate sacral neuromodulation for management of urinary and fecal incontinence in a pediatric population. Materials and Methods: This multicenter, open label, randomized, crossover study included children older than 5 years. After trial stimulation of the S3 root a neuromodulator (InterStim®) was implanted on the S3 foramen. Clinical examinations, voiding and bowel diaries, and urodynamic and manometric evaluations were performed at the beginning (t1) and end (t2) of the first period, and at the beginning (t3) and end (t4) of the second period. Results: A total of 33 patients (24 boys) with a mean ⫾ SD age of 12.22 ⫾ 5.09 years were randomized. Etiologies were mainly of neurological origin. Incontinence was mixed urinary and fecal in 19 cases, urinary only in 9 and fecal only in 5. Cystometric bladder capacity increased during sacral neuromodulation (delta ⫹24.27 ml vs ⫺37.45 ml, p ⫽ 0.01). There was no significant change in other urodynamic or manometric parameters. Overall positive response rate was more than 75% for urinary (81%) and bowel (78%) function. Crossover analysis indicated that sacral neuromodulation is more effective than conservative treatment for both types of incontinence (p ⫽ 0.001). Conclusions: In a pediatric population sacral neuromodulation is effective for bladder and bowel dysfunction and should be considered before irreversible surgery. Key Words: congenital abnormalities, electric stimulation therapy, fecal incontinence, lumbosacral plexus, urinary incontinence
696
www.jurology.com
SPHINCTER disorders can have many etiologies in children but the most frequent (95%) is congenital vertebral and medullary malformation, eg spinal dysraphism.1 Neurogenic bladder is characterized by loss of ability to control voiding. With age most patients have obstructive uropathy with increase in bladder pressure during filling and vesicosphincteric dyssyn-
ergia. Since the lower urinary and gastrointestinal tracts have the same embryological origin and innervations, sphincter disorders are frequently associated with fecal incontinence (80%).1,2 SNM has been used for management of sphincter disorders in adults.3–8 However, only 3 groups have reported results in children.4,9 –11 Most pediatric
0022-5347/10/1842-0696/0 THE JOURNAL OF UROLOGY® © 2010 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION
Vol. 184, 696-701, August 2010 Printed in U.S.A. DOI:10.1016/j.juro.2010.03.054
AND
RESEARCH, INC.
SACRAL NEUROMODULATION, AND URINARY AND FECAL INCONTINENCE
studies have involved small patient populations and data have been mainly clinical. In a previous study describing SNM in children with neurogenic bladder, we reported improvement in bladder and bowel function.9 These encouraging results led to the present study, which was designed to evaluate the efficacy and tolerance of SNM for management of urinary and fecal incontinence in a pediatric population.
METHODS Patients and Setting This prospective, national, randomized, open-label, crossover study was conducted in the pediatric surgery departments of French university hospital centers in Marseille, Lille, Besançon, Caen, Nancy, Paris, Lyon and Toulouse. Patients with urinary incontinence due to neurogenic bladder and/or fecal incontinence due to congenital malformation were eligible. Children or teenagers under surveillance for fecal and/or urinary incontinence were included if they presented with at least 2 of the following conditions— duration of continence 90 minutes or less, post-void residual volume greater than 50% of functional bladder capacity, bladder compliance less than 15 and bladder overactivity with pressure peaks exceeding 40 cm H2O. Ability and motivation of patient and family to comply with keeping a voiding and bowel diary, and attend followup examinations throughout the study were also taken into account for inclusion. All treatment, especially anticholinergic medication, was discontinued during the screening period. The main exclusion criteria were local
697
risk factors for device implantation (scar and sacral agenesis involving greater than 50% of the sacrum) and failure to detect the S3 roots. Before inclusion all patients underwent renal ultrasound, urethrocystography and renal scintigraphy (dimercapto-succinic acid) to evaluate bladder and kidney status. Routine preanesthetic laboratory tests (including creatininemia) were performed preoperatively. If clinical status of the patient changed during followup, testing was repeated to detect deterioration. Written informed consent was obtained from the parents and patients according to age. The study protocol was approved by a French ethics committee and registered as a clinical trial (AFSSAPS 2005/04/003) at the French Ministry of Health.
Implantation Technique As described previously, S3 roots were stimulated percutaneously with the patients under general anesthesia.9 If responses were acceptable, the neurostimulation device and lead were implanted immediately. The test stimulation period used in adults was not applied because we consider it too short to induce a response in children with neurological conditions. Patients were discharged home after 48 hours, provided that general status and pain level allowed.
Randomization A randomization list was generated by computer using a permuted block design. Subjects were randomly assigned to 1 of 2 neuromodulation groups, 6 months ON followed by 6 months OFF (group A) or the opposite sequence (group B). The 2 phases were separated by a 45-day washout period to return to baseline status (see figure).
Stimulation of S3 root
Positive Implantation Lead in S3 foramen, generator S/C
Negative excluded
Randomisation
Group A (Stimulation: ON)
Group B (Conservative treatment: OFF)
End therapeutic phase 1: 7th month “Cross Over” Beginning therapeutic phase 2: 9th month Group A (Conservative treatment: OFF) End therapeutic phase 2: 15th month Study design
Group B (Stimulation: ON)
698
SACRAL NEUROMODULATION, AND URINARY AND FECAL INCONTINENCE
Stimulation Neurostimulation was activated 24 hours after implantation in group A or after the OFF phase in group B. The initial pulse duration setting was 210 milliseconds in the cyclic mode (ON/OFF, 28/8 seconds). At each followup pulse amplitude (range 1 to 9 V) and frequency (10 to 20 Hz) were adjusted to maintain desired motor responses.
Data Collection At enrollment in the study (baseline evaluation, t1) demographic, clinical and severity data were collected for each patient. Following implantation response was evaluated at the end of the first 6-month period (t2), and at the beginning (t3) and end (t4) of the second period. At each evaluation physical examination was performed, including assessment of clinical status based on voiding and bowel diaries, and measurement of urodynamic and rectomanometric parameters.
End Points Patients were classified as responders or nonresponders regarding urinary and fecal performance after each phase of the ON and OFF sequence. Response was defined as resolution of urinary leakage and/or fecal soiling with no need for pads, or a decrease of more than 50% in the number of leaks and/or soilings with minimum protection needed between the beginning and end of each sequence. All other findings were scored as “no response.” The primary end point was defined by the overall response determined by combining clinical urinary and fecal responses. SNM was considered effective if either urinary or fecal response occurred and ineffective if neither urinary nor fecal response occurred. Secondary end points were urodynamic and rectomanometric parameters. In all patients the evaluation technique consisted of insertion of a triple lumen 8Fr to 10Fr Bohler urethral catheter, slow infusion of saline (20 ml per minute) and recording of abdominal pressure. Urodynamic assessment was performed in accordance with International Continence Society guidelines and included measurement of leak point pressure, cystometric bladder capacity, bladder activity, sphincter pressure, bladder compliance and maximal cystometric bladder capacity.12,13 Fecal assessment included the parameters transit time, rectal volume, resting pressure and voluntary contraction pressure. Changes in each parameter were compared between the end and beginning of the period (t2 vs t1 or t4 vs t3).
Statistical Analysis The sample size was designed to obtain 80% power to detect a 40% difference in clinical response at 6 months between the 2 groups, considering a potential loss to followup rate of 20% and S3 root detection failure of 20%. Comparison of gender, age and type of incontinence between groups A and B was performed using Student’s t test or Mann-Whitney test for quantitative variables and chi-square or Fisher’s exact test for frequencies. Primary and secondary outcomes were compared between the ON and OFF periods. The primary outcome (clinical response) was analyzed using the McNemar test. Secondary outcomes (urodynamic and fecal parameters) were analyzed using the Wilcoxon signed rank test. The second analyses included 2 subgroups, ie urinary incontinence (isolated
urinary or mixed incontinence) and fecal incontinence (isolated fecal or mixed incontinence). Statistical analysis was performed using SPSS®, version 15.0 software. Differences were considered significant at p ⫽ 0.05.
RESULTS Sample A total of 41 patients underwent trial assessment between April 2004 and September 2007. The S3 root was detected in only 33 patients who were randomized (overall implantation success was 81%). There were 24 boys and 9 girls with a mean ⫾ SD age of 12.22 ⫾ 5.09 years. Incontinence was urinary only in 9 patients, fecal only in 5 and mixed in 19. A total of 17 patients with urinary incontinence were on CIC. The underlying etiologies were spina bifida in 10 patients, sacral agenesis in 8, miscellaneous neurological anomalies in 7 (including 2 tumors), and congenital coloanal and urinary malformations in 5. Patients were randomly divided into 2 treatment groups, with 16 in group A and 17 in group B. Three patients randomized to group B were treated using the group A protocol due to a mix-up in instructions given to 1 team. The final population was 19 patients in group A and 14 in group B. At inclusion the 2 groups were comparable regarding gender, age and type of incontinence (table 1). A total of 24 patients completed the ON and OFF periods. Six patients dropped out due to device infection in 3, degradation of urinary incontinence following deactivation that required reactivation in 1, device removal and refusal to continue the trial in 1, and severe complications of the initial neurological pathology (Wolfram syndrome) in 1. In the remaining 3 patients clinical response was not available at the end of either the first or the second period. Efficacy of SNM Clinical response was significantly better when SNM was ON than OFF (75% vs 21%, p ⫽ 0.001). No patient was scored as a responder when SNM was OFF and nonresponder when SNM was ON. Respondership was higher for urinary and fecal incontinence when SNM was ON than OFF (81% vs 24%, p ⫽ 0.001 and 78% vs 17%, p ⫽ 0.001, respectively). In the crossover model using the McNemar test comparison of clinical responses showed “stimulation Table 1. Characteristics of randomized groups
M-to-F ratio Mean ⫾ SD age (yrs) No. incontinence: Fecal Mixed Urinary
Group A
Group B
Total
p Value
5:14 12.21 ⫾ 4.72
4:10 12.23 ⫾ 5.73
9:24 12.22 ⫾ 5.09
1.00 0.99 0.88
3 10 6
2 9 3
5 19 9
SACRAL NEUROMODULATION, AND URINARY AND FECAL INCONTINENCE
Table 2. Global clinical response between ON and OFF treatments OFF
No. ON Response (%)
Response No response Total
No. ON No Response (%)
Total No. (%)
5 13
0 6
5 (21) 19 (79)
18 (75)
6 (25)
24 (100)
⫽ response and control ⫽ no response” in 13 patients and “stimulation ⫽ no response and control ⫽ response” in none (p ⫽ 0.001, tables 2 and 3). The crossover model was also used to compare changes in urodynamic and manometric variables through time during each therapeutic phase. This analysis took into account the impact of period and sequence on therapeutic benefit. A significant increase in cystometric bladder capacity was observed during stimulation (delta ⫹24.27 ml vs ⫺37.45 ml, p ⫽ 0.01). The bladder was significantly more overactive with than without neuromodulation (⫹1 vs ⫺0.36, p ⬍0.001). No significant difference was noted between other urodynamic and rectomanometric variables (table 4). Tolerance of SNM The procedure was well tolerated. No patients dropped out of the study due to worsening urodynamic parameters with upper tract deterioration. Infection was observed in 4 patients (all nonresponders) and required removal after 1 month in 1, after the first ON phase in 2 and at the end of the trial in 1. Lead revision was required in 2 responders. Overall complication rate was 18% (6 of 33 patients).
DISCUSSION This study demonstrates that SNM can improve urinary and fecal continence in children with congenital malformations. SNM was introduced in 1981 for treatment of stress urinary incontinence or dysuria in adults. A benefit for anal incontinence was discovered almost by chance in adults treated for urinary disorders. Following this discovery SNM has also been proposed for isolated anal incontinence since 1995.3 In 2004 we reported subjective improvement in intestinal transit in children with neurological conditions.9 In a subsequent study aimed at assessing SNM for dysfunctional elimination syndrome in children without neurological conditions Roth et al reported improvement in urinary incontinence and constipation in 88% and 71% of patients, respectively.10 The present study revealed almost identical or better results for fecal incontinence (78%) but slightly lower results for urinary incontinence (81%). The latter finding is probably due to the neurological origin of urinary incontinence, which
699
Scheepens14 and Keppene15 et al consider an unfavorable predictor, especially for extensive lesions with long-standing symptomatology (eg congenital malformations). Lombardi et al obtained good results in partial spinal cord injury when the time between diagnosis and SNM was short.16 This finding suggests that early implantation may maximize benefits. The clinical improvement in urinary incontinence observed in this study was not always correlated with improvement in urodynamic parameters. Others have reported the same disconnect for neuromodulation,4 as well as other therapeutic treatments such as bulking agents in the bladder neck.17,18 Nevertheless, our crossover analysis comparing urodynamic variables after the 2 phases of treatment showed a statistically significant favorable change in cystometric bladder capacity under SNM. Our findings also demonstrated some bladder overactivity during SNM compared to conventional therapy, probably in relation to discontinuation of anticholinergic therapy. This result was not related to clinical outcome. Improvement in fecal incontinence observed in this study similarly was not always linked to manometric findings. Previous reports have been contradictory in this regard. Several authors have noted no manometric improvement in adults treated with SNM for fecal incontinence related or unrelated to neurological malformations,16,19,20 while others have reported improvement. 21–23 In this series intestinal transit time was prolonged with SNM but the change was not statistically significant. This result is in accordance with the findings of Michelsen et al, who reported that SNM may prolong colonic transit time by reducing the antegrade propagating sequence frequency.24 The physiological mechanisms underlying the mode of action of SNM on fecal and urinary incontinence are still barely understood. Clinical and experimental studies have revealed that SNM impacts bowel frequency and rectal sensation but has only low effect on the compliance of the anal sphincter.20,25–27 Boyle et al showed that the outcome of SNM for treatment of fecal Table 3. Clinical response between ON and OFF treatments in urinary and fecal incontinence subgroups OFF
No. ON Response (%)
Isolated or mixed urinary incontinence: Response 5 No response 12 Total 17 (81) Isolated or mixed fecal incontinence: Response 3 No response 11 Total p ⫽ 0.001 (McNemar).
14 (78)
No. ON No Response (%)
Total No. (%)
0 4
5 (24) 16 (76)
4 (19)
21 (100)
0 4
3 (17) 15 (83)
4 (22)
18 (100)
700
SACRAL NEUROMODULATION, AND URINARY AND FECAL INCONTINENCE
Table 4. Comparisons between ON and OFF stimulations for urinary and fecal parameters ON (mean ⫾ SD delta)* Urinary parameters: Leak point pressure Cystometric bladder capacity Overactivity Sphincter pressure Bladder compliance Total bladder capacity Fecal parameters: Transit time Rectal vol Resting pressure Voluntary contraction pressure
OFF (mean ⫾ SD delta)*
Order Effect
Time Effect
Treatment Effect
7.33 ⫾ 20.90 24.27 ⫾ 84.18 1.00 ⫾ 17.75 7.35 ⫾ 29.46 1.04 ⫾ 12.54 23.38 ⫾ 98.71
⫺5.42 ⫾ 32.64 ⫺37.45 ⫾ 77.85 ⫺0.36 ⫾ 24.82 ⫺5.50 ⫾ 46.36 ⫺5.30 ⫾ 16.56 ⫺20.04 ⫾ 94.30
0.74 0.04 0.08 0.28 0.59 0.02
0.73 0.59 0.48 0.59 0.39 0.81
0.31 0.01 ⬍0.001 0.52 0.31 0.14
17.38 ⫾ 40.28 ⫺15.00 ⫾ 29.58 8.37 ⫾ 25.11 6.35 ⫾ 28.81
⫺24.13 ⫾ 76.22 ⫺4.72 ⫾ 49.22 ⫺3.16 ⫾ 28.31 ⫺3.26 ⫾ 26.94
0.56 0.20 0.86 0.95
0.68 0.77 0.75 0.18
0.10 0.63 0.20 0.46
* Difference in values at beginning and end of each therapeutic period.
incontinence is not affected by the presence or severity of anal sphincter defects.28 Other research projects using functional and standard magnetic resonance imaging focus on the effects of SNM in patients with overactive bladder or urinary retention. Blok et al reported that in patients with urge urinary incontinence SNM seems to modulate areas in the brain normally involved in alertness and awareness, ie having a role in storage and voiding.29 In a recently published study Kavia et al concluded that in patients with urinary retention and Fowler syndrome SNM inhibits the urethral afferents to enable voiding.30 All of our patients undergoing CIC, even responders, still required CIC despite SNM. In most adult studies deactivating the neuromodulator led to reappearance of symptoms, at least initially. However, in a study of 20 children with elimination disorders Roth et al observed 2 patients who remained asymptomatic after deactivation.10 In our series of 33 patients 5 reported continued response after deactivation (table 1). These findings suggest partial neurological deficit and a possible role of SNM in training faulty neural pathways responsible for the insult. One of our patients presented with bladder overactivity following sacral agenesis. This patient noticed improvement without urodynamic correlation and wished to remain under SNM at the end of study. Urodynamic findings in this patient normalized 1 year after the study. This result suggests that clinical response occurs earlier than urodynamic re-
sponse. A study with longer followup is planned to address these issues. Use of SNM in growing children poses the problem of lead migration and battery longevity. The lead design offers the possibility of 4 different stimulation points, which is compatible with some growth. Battery replacement is simple. The complication rate in this study was 18.8%. Two types of complications occurred, ie infection (4 cases) and electrode migration (2). Surgical intervention to remove the device or reposition the electrode was required in all cases. Comparable complication rates have been observed in previous pediatric reports with no deleterious long-term consequences.10 There was no evidence that complications were more common according to type of incontinence. Device miniaturization and lead redesign for children should reduce future technical problems and decrease morbidity.
CONCLUSIONS Numerous studies have confirmed the efficacy of SNM in urinary and digestive disorders in adults. The present study demonstrates that SNM can also be effective for neurogenic incontinence in children. Based on these findings, we recommend attempting SNM before irreversible surgery.
ACKNOWLEDGMENTS Prof. P. Auquier and Mr. A. D. Loundou, Department of Public Health, Marseille Medical School, assisted in statistical analysis.
REFERENCES 1. Nijman RJ and Tekgul S: Pathophysiology of neurogenic bladder dysfunction. In: Pediatric Neurogenic Bladder Dysfunction—Diagnosis, Treatment, Long-Term Follow-Up. Edited by C Esposito, JM Guys, D Gough et al. New York: Springer 2006; chapt 4, p 33.
2. Ponticelli A, Iacobelli BD and Silveri M: Colorectal dysfunction and faecal incontinence in children with spina bifida. Br J Urol, suppl., 1998; 81: 117. 3. Matzel KE, Stadelmaier U and Hohenfellner M: Electrical stimulation of sacral nerves for treat-
ment of faecal incontinence. Lancet 1995; 346: 1124. 4. Bosch JL and Groen J: Sacral nerve modulation in the treatment of patients with refractory motor urge incontinence: long-terms results of a pro-
SACRAL NEUROMODULATION, AND URINARY AND FECAL INCONTINENCE
spective longitudinal study. J Urol 2000; 163: 1219. 5. Indar A, Young-Fadok T and Cornella J: A dual benefit of sacral neuromodulation. Surg Innov 2008; 15: 219. 6. Tjandra JJ, Lim JF and Matzel K: Sacral nerve stimulation: an emerging treatment for fecal incontinence. Aust N Z J Surg 2004; 74: 1098. 7. Fandel T and Tanagho EA: Neuromodulation in voiding dysfunction: a historical overview of neurostimulation and its application. Urol Clin North Am 2005; 32: 1. 8. Bernstein AJ and Peters KM: Expanding indications for neuromodulation. Urol Clin North Am 2005; 32: 59. 9. Guys JM, Haddad M, Planche D et al: Sacral neuromodulation for neurogenic bladder dysfunction in children. J Urol 2004; 172: 1673. 10. Roth TJ, Vandersteen DV, Hollatz P et al: Sacral neuromodulation for dysfunctional elimination syndrome: a single center experience with 20 children. J Urol 2008; 180: 306. 11. Hoebeke P, Van Laecke E, Everaert K et al: Transcutaneous neuromodulation for the urge syndrome in children: a pilot study. J Urol 2006; 166: 2416.
701
14. Scheepens WA, Jongen MM, Nieman FH et al: Predictive factors for sacral neuromodulation in chronic lower urinary tract dysfunction. Urology 2002; 60: 598.
22. Matzel KE, Stadelmaier U, Hohenfellner M et al: Chronic sacral spinal nerve stimulation for fecal incontinence: long-term results with foramen and cuff electrodes. Dis Colon Rectum 2001; 44: 59.
15. Keppene V, Mozer P, Chartier-Kastler E et al: Neuromodulation in the management of neurogenic lower urinary tract dysfunction. Prog Urol 2007; 17: 609.
23. Ripetti V, Caputo D, Ausania F et al: Sacral nerve neuromodulation improves physical, psychological and social quality of life in patients with fecal incontinence. Tech Coloproctol 2002; 6: 147.
16. Lombardi G, Del Popolo G, Cecconi F et al: Clinical outcome of sacral neuromodulation in incomplete spinal cord-injured patients suffering from neurogenic bowel dysfunctions. Spinal Cord 2009; 11: 101. 17. Guys JM, Breaud J, Hery G et al: Endoscopic injection with polydimethylsiloxane for the treatment of pediatric urinary incontinence in the neurogenic bladder: long-term results. J Urol 2006; 175: 1106. 18. Lottmann HB, Margaryan M, Bernuy M et al: The effect of endoscopic injections of dextranomer based implants on continence and bladder capacity: a prospective study of 31 patients. J Urol 2002; 168: 1863. 19. Faucheron JL, Bost R, Duffournet V et al: Sacral neuromodulation in the treatment of severe anal incontinence. Forty consecutive cases treated in one institution. Gastroenterol Clin Biol 2006; 30: 669.
12. Abrahams P, Blaivas JG, Stanton SL et al: The standardisation of terminology of lower urinary tract function. Scand J Urol Nephrol 1998; 114: 5.
20. Uludag O, Koch SM, Van Gemert WG et al: Sacral neuromodulation in patients with fecal incontinence: a single-center study. Dis Colon Rectum 2004; 47: 1350.
13. Everaert K, Plancke H, Lefevere F et al: The urodynamic evaluation of neuromodulation in patients with voiding dysfunction. Br J Urol 1997; 79: 702.
21. Leroi AM, Michot F, Grise P et al: Effect of sacral nerve stimulation in patients with fecal and urinary incontinence. Dis Colon Rectum 2001; 44: 779.
24. Michelsen HB, Christensen P, Krogh K et al: Sacral nerve stimulation for faecal incontinence alters colorectal transport. Br J Surg 2008; 95: 779. 25. Dinning PG, Fuentealba SE, Kennedy ML et al: Sacral nerve stimulation induces pan-colonic propagating pressure waves and increases defecation frequency in patients with slow-transit constipation. Colorectal Dis 2007; 9: 123. 26. Uludag O, Morren GL, Dejong CH et al: Effect of sacral neuromodulation on the rectum. Br J Surg 2005; 92: 1017. 27. Vitton V, Gigout J, Grimaud JC et al: Sacral nerve stimulation can improve incontinence in Crohn’s patients with internal and external sphincter disruption. Dis Colon Rectum 2008; 51: 924. 28. Boyle DJ, Knowles CH, Lunniss PJ et al: Efficacy of sacral nerve stimulation for fecal incontinence in patients with anal sphincter defects. Dis Colon Rectum 2009; 52: 1234. 29. Blok BF, Groen J, Bosch JL et al: Different brain effects during chronic and acute sacral neuromodulation in urge incontinent patients with implanted neurostimulators. BJU Int 2006; 98: 1238. 30. Kavia R, Dasqupta R, Critchley H et al: A functional magnetic resonance imaging study of the effect of sacral neuromodulation on brain responses in women with Fowler’s syndrome. BJU Int 2010; 105: 366.
EDITORIAL COMMENT The authors are to be congratulated for this original study. One must remember that few studies are designed for the pediatric population, whatever the topic. The authors had the novel idea of focusing on global pelvic floor dysfunction (urinary and fecal incontinence), which increases the interest of the study. In a time of evidence-based medicine this is probably the only controlled and well shaped study to evaluate the effect of SNM on a neurogenic pediatric population. Despite the fact that SNM may or may not be effective at improving quality of life and urinary incontinence in an adult neurogenic popula-
tion,1 it is well-known that it can demonstrate individual usefulness. SNM is a conservative therapy that may act on urinary and fecal incontinence. New treatments are emerging for neurogenic urinary incontinence,2 provided that patients are using clean intermittent catheterization to empty the bladder. This series gives us a more clear and independent evaluation of SNM in a unique study design. Emmanuel Chartier-Kastler Department of Urology Pierre and Marie Curie Faculty of Medicine Paris, France
REFERENCES 1. Lombardi G and Del Popolo G: Clinical outcome of sacral neuromodulation in incomplete spinal cord injured patients suffering from neurogenic lower urinary tract symptoms. Spinal Cord 2009; 47: 486. 2. Game X, Mouracade P, Chartier-Kastler E et al: Botulinum toxin-A (Botox) intradetrusor injections in children with neurogenic detrusor overactivity/neurogenic overactive bladder: a systematic literature review. J Pediatr Urol 2009; 5: 156.