Functional Nonretentive Fecal Incontinence: Do Enemas Help?
RosaBurgers, MD1, JohannesB.Reitsma, MD2, MarloesE.J. Bongers,MD1, Fleurde Lorijn, MD1, andMarc A.Benninga,MD1
Objective To assess the current treatment of functional nonretentive fecal incontinence, which consists of edu-
cation, toilet training, and positive motivation.
Study design Patients, age 6 years and older, referred for fecal incontinence (FI) and diagnosed with functional
nonretentive fecal incontinence were eligible candidates. Seventy-one children (76% boys, median age 9.3 years)
were randomized to receive conventional therapy (control group) or conventional therapy in addition to daily
enemas during 2 weeks. Treatment success was defined as <2 episodes of FI/month without use of enemas.
Results At intake, the median FI frequency was 6.1 per week, whereas the median defecation frequency was 7.0
per week. At the end of the treatment period, the median number of FI episodes was significantly decreased in both
groups: from 7.0 (IQR 4.0-11.5) to 1.0 (IQR 0.5-2.0) in the intervention group and from 6.0 (IQR 4.0-10) to 2.0 (IQR
0.5-3.5) in the control group. No statistical difference was found between the groups at the end of the treatment
period (P = .08) nor during additional follow-up (average success rate 17% for both groups, P = .99).
Conclusion Temporarily application of additional rectal enemas did not significantly improve treatment success
compared with conventional therapy alone. (J Pediatr 2013;162:1023-7).
constipation, and children are treated with laxatives (oral and enemas).2-4The remaining 10% of patients present with FI as
a single symptom without any organic cause or sign of constipation and is currently classified as functional nonretentive fecal
randomized controlled trial in children with FNRFI showed that oral laxatives are not helpful and even increase the number of
change abnormal defecation dynamics but this had no effect on treatment success.7Anecdotal success has been described using
loperamide in childhood-onset, long-standing FNRFI.8
Despite intensive treatment programs, success rates after 2 years of medical and/or behavioral treatment were only 29% in
children with FNRFI. Moreover, at the age of 12 years, almost 50% of the children still suffered from FI.9It has been suggested
that children with FNRFI deny or neglect normal physiological stimuli to defecate and contract the external anal sphincter to
retain stool in the rectum.2We hypothesized that the daily use of rectal enemas, which results in clean underwear, can lead to
daily enemas in addition to conventional therapy would increase the numberof patients with treatment success compared with
conventional therapy alone.
ecal incontinence (FI) represents an upsetting and psychologically distressing problem in children and results in lower
health-related quality of life.1In approximately 95% of children with FI, no organic cause can be identified; these chil-
dren are considered to have a functional defecation disorder. In approximately 90% of these children, FI is the result of
Tobe eligibleforthetrial,patientshadtofulfillallofthefollowingcriteria:presence ofFIwith>1episodeofFIperweekduring
in the rectum at physical examination, and age between 4 and 17 years.6,7Children with stool withholding behavior were con-
sidered as eligible candidates for this study. Children with a delayed colonic transit time (CTT) and organic causes of FI such as
muscle disorders, spina bifida, and children operated for anal atresia or Hirschsprung’s disease, and those children with mental
retardation were excluded.
All children were referred for functional defecation disorders by general prac-
titioners, school doctors, and pediatricians to a tertiary hospital (Emma Chil-
dren’s Hospital/AMC, Amsterdam, The Netherlands). All parents and
From the1Department of Pediatrics, Emma Children’s
Hospital, Amsterdam Medical Center; and2Department
of Clinical Epidemiology, Biostatistics and
Bioinformatics, Academic Medical Center, Amsterdam,
The authors declare no conflicts of interest.
Study was registered with The Netherlands National Trial
Register: NTR 65 and ISRCTN28937219.
0022-3476/$ - see front matter. Copyright ª 2013 Mosby Inc.
All rights reserved. http://dx.doi.org/10.1016/j.jpeds.2012.10.037
Colonic transit time
Functional nonretentive fecal incontinence
Generalized estimating equations model
children >12 years gave written informed consent. The study
protocol was approved by the medical ethics committee of
Children were randomized on a 1:1 basis to one the follow-
or conventional treatment combined with regular application
of rectal enemas (intervention group). Randomization was
carriedoutwith thehelpofa specialized programon a central
computer. The randomization was stratified by sex.
To obtain baseline data assessment, children were asked to
record, with assistance from their parents, their defecation
pattern (defecation frequency and episodes of FI) in a daily
diary a week before randomization (t = ?1 until t = 0). In
all patients, medication influencing motility was discontin-
ued during this week.
Before randomization, each child underwent a complete
and rectal examination, and CTT measurement. A standard-
ized interview was conducted to determine whether children
fulfilled the criteria for FNRFI.10The child and parents were
asked about bowel function, frequency of defecation and FI,
consistency and size of stool, pain during defecation, and
associated symptoms such as abdominal pain, appetite, and
urinary incontinence. The standardized questionnaire at in-
take also included questions regarding medical history, age
at onset of defecation problems, and laxative use.
Abdominal examination focused on distension and the
presence of an abdominal fecal mass. Rectal examination in-
cluded inspection for fissures, hemorrhoids, scars, gluteal
cleft deviation, or dimples. Digital examination was per-
formed to provide information about anal tone and the pres-
ence of a rectal fecal mass.
han et al.12A single abdominal radiograph was taken follow-
ing ingestion of capsules, each containing 10 radio-opaque
markers (Sitzmarks; BioPharma, Weesp, The Netherlands),
on 6consecutivedays.11This procedurewasperformedwith-
out bowel preparation. Calculation of CTT was performed
according to a previously described formula (CTT in hour-
s = number of markers ? 2.4). When the total CTT exceeded
62 hours, it was considered to be delayed, indicating consti-
pation.12Subsequently, the patient was excluded from the
Patientswererandomizedto 1ofthe 2treatment groups.The
controlgroup wastreated according toconventional therapy;
education, toilet training, behavioral strategies, bowel diary,
about the different aspects of FI, with an explicit effort to al-
leviate guilt and to be nonaccusatory. All children were in-
structed to try to defecate on the toilet for 5 minutes after
each meal. Motivation was enhanced by praises and small
gifts. After the 6-week active treatment period, patients in
the control group were allowed to start the use of enemas.
The intervention group was treated with rectal enemas in
addition to the conventional therapy as described above.
Standardized instructions were given to the parentsand child
regarding the administration of the rectal enemas. Children
were instructed to bend their knees while lying sidewards
on their bed and attempt to defecate during introduction
of the enema. Furthermore, we recommended to heat the en-
to retain the fluid until strong urge to defecate was felt.
During the first 2 weeks of the 6-week active treatment
period, patients were instructed to use daily enemas (dioctyl-
sulfosuccinate sodium, Klyx; [Pharmachemie, Haarlem, The
Netherlands] 120 mL for children $6 years of age or
natriumfosfaat; Colex Klysma; [Tramedico BV] 133 mL for
children $6 years of age) at home. Thereafter, this frequency
was reduced stepwise by 1-2 per week every week if the phy-
sician considered treatment successful. Success was defined
as a 50% reduction in the number of FI episodes compared
with number of FI episodes occurring the week before.
Measurements of Outcome
All patients had outpatient visits/telephone contacts during
which therapy compliance and defecation frequency and FI
episodes (information from bowel diary) were discussed at
1, 2, 4, and 6 weeks after the start of the treatment. Children
receiving conventional therapy with persistence of episodes
of FI were able to start with rectal enemas after the 6-week
active treatment period. Two more assessments were per-
defecation pattern and use of medication during additional
follow-up. Follow-up was performed either during a clinical
visit or by phone with a standard questionnaire.
The primary outcome measures were FI frequency per
week and treatment success during active treatment and ad-
ditional follow-up. Treatment was considered successful if
the child had less than 2 episodes of FI/month without use
ofenemas. Secondaryoutcomemeasures weredefecation fre-
quency per week, painful defecation, and abdominal pain
during defecation. Clinical improvement was defined as #1
FI/week regardless the use of enemas.
To study outcome measures over time, we differentiated
theseries of visits in 2periods; the first6 weeks after random-
visits performed at 12 weeks, 6 months, and 1 year after ran-
domization considered as the ‘additional follow-up period’.
Sample Size Consideration
Before start of the study, a sample size calculation was made,
expecting a 30% difference in the proportion of success
between the control and intervention treatment. To detect
such a difference using a 2-sided significance level of 0.05
with a power of 0.80 would require a sample size of 39 chil-
dren in each group.
The incidence of adverse events during the weeks of treat-
ment with enemas was also documented. Both the incidence
and severity of gastrointestinal adverse events (abdominal
pain, painful defecation) were recorded in the diary and
THE JOURNAL OF PEDIATRICS
Vol. 162, No. 5
Burgers et al
assessed at weeks 1, 2, 4, and 6 of the period of the trial on
a 3-point scale: 0 = no gastrointestinal symptoms present,
Demographic and clinical characteristics at baseline were an-
according to an intention-to-treat approach. Outcome data
consisted of repeated measurements over time in the same
patient; therefore, we applied models that explicitly take
into account the correlation that is likely to exist between
measurements within the same individual. To study the suc-
cess rates over time, we used the generalized estimating equa-
tions (GEE) model. GEE models are used to analyze trends
over time in binary outcomes (for this study; treatment suc-
cess). GEE models are an extension of generalized linear
models to deal with correlated outcomes.13Within the GEE
framework, a working correlation matrix is estimated to ad-
just the standard parameters for the correlation that is pres-
ent. The working correlation matrix was considered
exchangeable. The GEE models contained the following vari-
ables: treatment given, time as a categorical variable, and the
interaction between treatment and time. Results are pre-
sented as OR together with 95% CIs.
Statistical analyses were performed by using SPSS Win-
dows v. 16.0 (SPSS Inc, Chicago, Illinois). Statistical signifi-
cance was accepted at P < .05.
Between November 2000 and December 2009, a total of 71
children were enrolled fulfilling the Rome II criteria for
FNRFI. Of these children, 17% had been seen only by their
general practitioner, 62% visited a general pediatrician, and
39% had visited a psychologist or psychiatrist. A total of 36
patients were allocated to the intervention group and 35 to
the control group. Unfortunately, enrollment of 39 children
per group, as aimed based on the sample size calculation, was
not reachedin 10 years. The baseline characteristics per treat-
ment group are presented in Table I. None of the patients in
the control group used enemas during the 6-week active
these children decreased the use of rectal enemas in advance
of the predetermined scheme. At the end of the 6-week treat-
ment period, 47% of the children in the intervention group
still used 5 enemas/week, whereas 42% of the children used
2-4 enemas per week. Only 2 children were successfully
treated at the end of the 6-week treatment period. These chil-
dren had no FI without the use of enemas.
Main Outcome Measures
The results of the intention-to-treat analysis, based on all 71
patients, with regards to the FI frequency per week and treat-
ment success (defined as successfulif the child had less than 2
episodes of FI/month without use of enemas) are shown in
Table II and the Figure. From baseline to the end of the
treatment period, the median number of FI episodes
significantly decreased from 7.0 (IQR 4.0-11.5) to 1.0 (IQR
0.5-2.0) in the intervention group and from 6.0 (IQR 4.0-
10) to 2.0 (IQR 0.5-3.5) in the control group. No statistical
difference was found between the 2 groups at the end of
the active treatment period (t = 6, P = .08) (Figure). No
significant difference (P = .3) was found in the number of FI
episodes at additional follow-up between the intervention
group and control group (median 1.6 [IQR 0.6-2.8] and
2.1 [IQR 1.1-3.1]), respectively. The number of FI episodes
during additional follow-up were statistically significantly
lower compared with the start of the treatment period in
both study groups (P < .001). The treatment success was
not significantly different between the groups during
additional follow-up (summary OR across all time points
1.0 [95% CI 0.4-2.4]; P = .99) with a mean success rate of
17% for both groups (Table II). The number of patients
fulfilling thecriteria for
significantly higher in the intervention group compared
with the control group during the active treatment period
(OR 2.1 [95% CI 1.1-3.9], P = .02) but not during
additional follow-up (OR 1.3 [95% CI 0.64-2.43], P = .51)
In the intervention group 50%, 25%, and 2.9% of the chil-
dren were still using enemas at 12, 26, and 52 weeks, respec-
which resulted in disappearance of FI, whereas 1 control
patient used enemas at t = 52, which resulted in a decrease
Table I. Demographic and clinical characteristics at
Intervention groupControl group
Number of patients
Median age in y (range)
Median defecation frequency/wk
Median FI episodes/wk
Pain during defecation
Passage of large stools
Use of laxatives
Urinary incontinence (day)
Urinary incontinence (night)
75 (n = 27)
33 (n = 12)
2.8 (n = 1)
70 (n = 25)
28 (n = 10)
86 (n = 31)
2.8 (n = 1)
25 (n = 9)
36 (n = 13)
77 (n = 27)
29 (n = 10)
8.6 (n = 3)
60 (n = 21)
23 (n = 8)
74 (n = 26)
37 (n = 13)
34 (n = 12)
*Depicted as proportion (%) or median (IQR).
Table II. Treatment success and clinical improvement
OR (95% CI)P value
Functional Nonretentive Fecal Incontinence: Do Enemas Help?
in the number of FI episodes from 5 at intake to 1 FI episode/
week at the end of the follow-up period.
(IQR 7.2-14) in the control group (P = .5). During additional
follow-up the defecation frequency per week was 8.0 (IQR
6.5-11) in the intervention group and 8.9 (IQR 6.7-15) in the
control group (P = .6).
At intake, 3 children in the control group (8.6%) and 1
child in the enema group reported hard stools (P = .5).
The consistency of stools did not change during the interven-
tion period in both groups.
Abdominal pain during defecation was reported by only
a minority (2.8%-8.3%) of the children in the intervention
group during the active treatment period at different visits.
None of the patients in the control group reported abdomi-
nal pain during defecation. Urinary incontinence during day
and night decreased from 31% and 35% at intake to 27% and
7.0%, respectively, at the end of the intervention/active treat-
ment period (t = 6), and 9.8% and 17%, respectively, at 1-
year follow-up in the entire study cohort.
Pain during defecation in the anorectal region was reported
by 8.3%-17% in the intervention group during the active
treatment period, and in 5.6%-8.3% during additional
follow-up. In the control group, pain during defecation
was reported by 2.9%-17% during active treatment, and by
0%-8.6% during additional follow-up. None of the children
in the intervention group stopped the use of enemas because
of water and electrolyte disturbances.
This randomized controlled trial investigated the role of rec-
tal enemas in the treatment of children with FNRFI. A signif-
the intervention group and control group after the active
treatment period and during follow-up. However, rectal
enemas were not more effective than conventional therapy
in these children. Using strict success criteria, low success
rates were found in both treatment groups during follow-up.
The median number of FI episodes at intake and the de-
crease of the number of FI episodes after active treatment
in our current study is in line with our former randomized
controlled trial evaluating the effect of biofeedback training
in children with FNRFI.6It is noteworthy, however, that in
other randomized controlled trials in children with func-
tional gastrointestinal disorders, a significant improvement
in symptoms is observed during the first weeks of treat-
ment.14,15Most of these large clinical trials are performed
in tertiary centers. One might speculate that these experi-
enced academic pediatric gastroenterologists in the field of
functional gastrointestinal disorders arewellaware of the im-
portance of education, reassurance, and allocation of time to
the initial consultation to validate the patients’ symptoms
and respond to the patient’s specific needs and fears, might
influence the initial treatment effect. During follow-up, the
number of FIepisodes were stillsignificantlylowercompared
with intake in both groups, underlining the importance of
long-term follow-up of these patients. During the frequent
outpatient visits, we enhance motivation and stress the im-
portance of adherence to the treatment strategies.
Clinical improvement, defined as #1 FI/week regardless
in the intervention group compared with the control group
during the active treatment period but this difference is not
expanding throughout additional follow-up (OR 1.3). We
hypothesize that the fast reduction in the number of rectal
enemas per week, as put forward in the treatment protocol,
was too optimistic. These children might be better treated
when the frequency of enema application is maintained for
a period of at least 3 months and not reduced within the first
weeks. The extension of the treatment period with rectal en-
emasseems suitable becausewehave shownthat children tol-
erate rectal therapy surprisingly well without any side
effects.16,17A recent systematic review described the adverse
effects of sodium phosphate enema in adults and children.18
All side effects described were due to water and electrolyte
disturbances. The main risk factors were children younger
than 2 years of age and associated comorbidity.
Another explanation is that children are obliged to stick to
the toilet training and behavioral strategies with frequent use
of rectal enemas and tend to lose interest, trust, and compli-
ance when enemas are reduced. This seems plausible because
children with constipation considered rectal enemas impor-
tant to solve their defecation disorders.16
A high percentage of children was found with daytime and
night time urinary incontinence.6These rates, 31% for day-
time urinary incontinence and 35% for night-time urinary
incontinence, were much higher than reported in children
with constipation with or without FI seen in primary pediat-
ric clinics. In this group of constipated patients, only 3.3%
exhibited daytime, 1.8% day and night-time, and 5.4%
night-time urinary incontinence respectively.19It has been
suggested that early intervention with laxatives may prevent
fecal and/or urinary incontinence.19Nevertheless, the exact
Figure. Median FI episodes per week.
THE JOURNAL OF PEDIATRICS
Vol. 162, No. 5
Burgers et al
mechanism(s) underlying the relationship between defeca-
tion disorders and lower urinary tract symptoms remain un-
clear. We hypothesize that children with FNRFI not only
denyor neglect their urge todefecate butexhibitthesame be-
havior towards micturation.2
a significant and sustained decrease in the occurrence of uri-
nary incontinence.6This supports the idea that a strict regi-
men of toilet training together with demystification and
motivation is able to improve both urine and FI.20,21
A limitation of this study is that no questionnaires were
used to evaluate subjective feelings about and experience
with the application of rectal enemas for a longer period of
time. A recent study, however, showed that the prolonged
use of enemas in children with long-lasting constipation
was well tolerated.16The majority of these children (76%)
never/seldom felt worse after application of an enema,
whereas 15% perceived them as extremely terrible. Another
drawback of this studyis that wedidnot succeed in including
start of the study, despite our 10-year period of inclusion.
However, the identical success rate of 17% in both treatment
arms during additional follow-up makes it highly unlikely
that further inclusion of patients would have generated clin-
ically relevant differences.
In summary, temporary use of daily rectal enemas in addi-
tion to conventional therapy is not more effective than con-
ventional therapy alone in children with FNRFI. More
importantly, low success percentages were found after 1
year follow-up. Because the use of enemas is not more effec-
tive than conventional treatment, new treatment strategies
are necessary to successfully treat this group of children. n
The authors thank Chris J.C. Hoppenbrouwes, Wieger P. Voskuijl,
Maartje M. van den Berg, Noor L. Bekkali, Olivia Liem, Babette Pee-
ters, Clara M Loots, and Michiel P van Wijk for their substantial con-
tributions to the conception and acquisition of data.
Submitted for publication May 15, 2012; last revision received Jul 31, 2012;
accepted Oct 11, 2012.
Reprint requests: Rosa Burgers, MD, Department of Pediatric
Gastroenterology and Nutrition, Emma Children’s Hospital/Academic Medical
Center, Meibergdreef 9, C2-312, 1105 AZ Amsterdam, The Netherlands. E-
1. Bongers MEJ, van Dijk M, Benninga MA, Grootenhuis MA. Health
related quality of life in children with constipation-associated fecal
incontinence. J Pediatr 2009;154:749-53.
2. Benninga MA, B€ uller HA, Heymans HS, Tytgat GN, Taminiau JA. Is en-
copresis alwaysthe result of constipation?ArchDis Child 1994;71:186-93.
3. Dilorenzo C. Pathophysiology of pediatric fecal incontinence. Gastroen-
4. Bongers MEJ, Tabbers MM, Benninga MA. Functional nonretentive fe-
cal incontinence in children. J Pediatr Gastroenterol Nutr 2007;44:5-13.
5. Rasquin A, Di Lorenzo C, Forbes D, Guiraldes E, Hyams JS, Staiano A,
et al. Childhood functional gastrointestinal disorders: child/adolescent.
6. van Ginkel R, Benninga MA, Blommaart PJ, van der Plas RN,
Boeckxstaens GE, B€ uller HA, et al. Lack of benefit of laxatives as adjunc-
tive therapy for functional nonretentive fecal soiling in children. J Pe-
7. van der Plas RN, Benninga MA, Redekop WK, Taminiau JA, B€ uller HA.
Randomized trial of biofeedback training for encopresis. Arch Dis Child
8. Voskuijl WP, Ginkel RV, Taminiau JA, Boeckxstaens GE, Benninga MA.
Case report loperamide suppositories in an adolescent with childhood-
onset functional nonretentive fecal soiling. J Pediatr 2003;198-200.
9. Voskuijl WP, Reitsma JB, van Ginkel R, B€ uller HA, Taminiau JAJM,
Benninga MA. Longitudinal follow-up of children with functional non-
retentive fecal incontinence. Clin Gastroenterol Hepatol 2006;4:67-72.
10. VoskuijlWP, Heijmans J,Heijmans
applicability in clinical and research practice. J Pediatr 2004;145:213-7.
11. Bouchoucha M, Devroede G, Arhan P, Strom B, Weber J, Cugnenc PH,
et al. What is the meaning of colorectal transit time measurement? Dis
Colon Rectum 1992;35:773-82.
12. Arhan P, Devroede G, Jehannin B, Lanza M, Faverdin C, Dornic C, et al.
Segmental colonic transit time. Dis Colon Rectum 1981;24:625-9.
13. Diggle PJ, Liang KY, Zeger SL. Analysis of longitudinal data. Oxford:
Clarendon Press; 1994.
ter, randomized, placebo-controlled trial of amitriptyline in children with
functional gastrointestinal disorders. Gastroenterology 2009;137:1261-9.
15. Voskuijl W, de Lorijn F, Verwijs W, Hogeman P, Heijmans J, M€ akel W,
et al. PEG 3350 (Transipeg) versus lactulose in the treatment of child-
hood functional constipation: a double blind, randomized controlled,
multicenter trial. Gut 2004;53:1590-4.
16. Bongers MEJ, van den Berg MM, Reitsma JB, Voskuijl WP,
Benninga MA. A randomized controlled trial of enemas in combination
with oral laxative therapy for children with chronic constipation. Clin
Gastroenterol Hepatol 2009;7:1069-74.
17. Bekkali NLH, van den Berg MM, Dijkgraaf MGW, van Wijk MP,
Bongers MEJ, Liem O, et al. Rectal fecal impaction treatment in child-
hood constipation: enemas versus high doses oral PEG. Pediatrics
18. Mendoza J, Legido J, Rubio S, Gisbert JP. Systematic review: the adverse
19. Loening-Baucke V. Prevalence rates for constipation and fecal and uri-
nary incontinence. Arch Dis Child 2007;92:486-9.
20. Allen HA, Austin JC, Boyt MA, Hawtrey CE, Cooper CS. Initial trial of
timed voiding is warranted for all children with daytime incontinence.
21. Thom M, Campigotto M, Vemulakonda V, Coplen D, Austin PF. Man-
agement of lower urinary tract dysfunction: a stepwise approach. J Pe-
diatr Urol 2012;8:20-4.
Functional Nonretentive Fecal Incontinence: Do Enemas Help?