- Email: [email protected]
Balloon defecation as a predictor of outcome in children with functional constipation and encopresis
Balloon defecation as a predictor of outcome in children with functional constipation and encopresis Vera Loening-Baucke, MD From the Department of Pediatrics, Universityof Iowa, Iowa City
Objective: To evaluate whether the ability to defecate a rectal balloon might predict 12-month recovery in children with functional constipation and encopresis. Methods: We evaluated the ability to defecate within 5 minutes a 100 ml waterfilled rectal balloon by 20 healthy children and 139 children with functional constipation and encopresis. Results: All healthy children and only 47% of the patients were able to defecate the balloon. Twelve months after the start of treatment, 5 I% of patients able to and 34% of patients unable to defecate the balloon had recovered (p <0.03). Logistic regression revealed that the ability to defecate the balloon and a history of secondary encopresis were related to recovery (p <0.04). Patients who were unable to defecate the balloon or who did not recover had significantly more impairment in anorectal functions than those who were able to defecate the balloon or who did recover. The ability of the balloon defecation test to predict recovery had a sensitivity of 57%, a specificity of 60%, a positive predictive value of 0.51, and a negative predictive value of 0.66. Conclusion: Children with functional constipation and encopresis who were able to defecate the rectal balloon were twice as likely to recover. Even though there was a clinically significant difference in the recovery rates between patients who could and those who could not defecate the balloon, calculation of predictive values showed that the balloon defecation test could not reliably predict recovery. (J PEDIATR1996; 128:336-40)
Constipation with or without encopresis (any amount of fecal soiling in children with stool retention) represents a common problem in children, accounting for 3% of visits to the University of Iowa Hospitals and Clinics Pediatric OutPresented in part at the 95th Annual Meeting of the American Gastroenterological Association, May 15, 1995, and published in abstract form (Gastroenterology 1995;108:640A). Supported by grant No. M01-RR-00059 from the General Clinical Research Center Program, Division of Research Resources, National Institutes of Health, and the Children's Miracle Network Telethon. Reprint requests: Vera Loening-Baucke, MD, University of Iowa Hospitals and Clinics, Department of Pediatrics, JCP 2555, 200 Hawkins Dr., Iowa City, IA 52242. Copyright © 1996 by Mosby-Year Book, Inc. 0022-3476/96/$5.00 + 0 9•20•69851
336
patient Clinic and to a large general pediatric clinic in Boston, ~ and for 25% of visits to a pediatric gastroenterology clinic.2 Constipation in children can be defined by a stool frequency of less than 3 per week; constipation can also be defined by the presence of a large rectal stool mass (stool retention) with or without encopresis, even when the stool frequency is three or more per week. The ability to defecate depends on several factors, including the simultaneous actions of the voluntary external and the involuntary internal anal sphincters, the tonicity of the rectum, and the condition of the spinal cord. Some children with functional constipation and encopresis have ineffectual straining, and up to 50% contract rather than relax the external anal sphincter and pelvic floor during straining for defecation (termed anismus, abnormal defecation dynamics or patterns, rectoanal or rectosphincteric dyssynergia, and
The Journal o[ Pediatrics Volume 128, Number 3
spastic pelvic floor syndrome3-8). Others have a megarecturn, rectal atony, or both, which hinder the propulsion of stool from the rectum by peristalsis.9-12 These abnormalities impede defecation. A method to test the coordinated act of defecation is the rectal balloon defecation test. 1315 The aim of our study was to evaluate whether the ability to defecate, as demonstrated by the defecation of a 100 ml water-filled rectal balloon, might predict 12-month recovery in children with functional constipation and encopresis.
METHODS Study patients. Two hundred thirty-two consecutive patients (176 boys and 56 girls, 5 to 18 years of age, mean age 9 + 3 years) with functional constipation and encopresis (one or more soiling episodes per week) were examined in the University of Iowa Encopresis Clinic with the balloon defecation test between 1985 and 1993. One hundred thirtynine of them (103 boys and 36 girls, mean age 10 _+ 3 years) were reexamined 12 months after the start of treatment and constitute the study group. These patients were healthy except for problems related to constipation and encopresis. Children with Hirschsprung disease, hypothyroidism, mental deficiency, chronic debilitating diseases, or neurologic abnormalities, and children who had previous surgery of the colon were excluded. Thirty-two healthy children (16 boys and 16 girls, 6 to 16 years of age, mean age 11 + 3 years) served as control subjects. All were studied with anorectal manometry, and 20 of them underwent the balloon defecation test.
Anorectal studies Studies of balloon defecation. To simulate defecation of a stool from the rectum, we asked children to defecate a 100 ml water-filled rectal balloon while sitting on a toilet chaff', allowing 5 minutes for defecation. The rectal balloon was made from a latex party balloon (Latex Occidental, distributed by American Imports, Inc., San Antonio, Tex.) and was 2.5 x 3 cm when deflated and 5 x 5 cm in size when filled with 100 ml of water. The empty balloon was lubricated and inserted into an empty rectum. AnorectaI manometry. In addition to the balloon defecation test, anorectal functions were assessed with anorectal manometry in all children. The methods were reported previously.lll 3 We determined the threshold volumes (measured in milliliters of air) required to elicit a transient sensation of rectal balloon distention, an initial urge to defecate, a strong urge to defecate (critical volume), a sustained complete relaxation of the internal and external anal sphincters (constant relaxation), and rectal contractility (10 mm Hg or more, recorded 5 cm below the base of the rectal balloon) by stepwise addition of 10 ml air increments, up to 60 ml, and then 30 ml increments, each 10 to 15 seconds into the rectal bal-
Loening-Baucke
337
loon.3, 4, 11 When the critical volume was reached, but constant relaxation or rectal contractility had not yet occurred, the next higher volume was used as the volume of constant relaxation or rectal contractility. While lying in the left lateral position, the child was asked to bear down five times as if defecating and to squeeze (tighten up) five times in random order. If no increase in intraabdominal pressure (rectal pressure) during the defecation trials was recorded, the child was instructed to increase intraabdominal pressure. We defined defecation dynamics as normal if the integrated electromyogram and the anal pressure decreased during increased intraabdominal pressure in two or more of five defecation trials. Anismus was defined by the presence of a persistent increase in the integrated electromyogram and the anal pressure during bearing down in four or more of five defecation trials,y' 8, 13 Treatment. All patients received conventional treatment for constipation and encopresis similar to that used for the past 15 years in our encopresis clinic. 16 The treatment included disimpaction, education to relieve the social and psychologic problems faced by these children and their parents, prevention of reaccumulation of fecal material in the colon with the help of daily laxative use and an increase in dietary fiber, and scheduled toilet sittings. 16 Parents and children were instructed to keep a diary of bowel movements, fecal soiling, and medication used. Forty-eight patients with anismus received, in addition to conventional treatment, at least two and up to six biofeedback training sessions (mean 3 + 2). The equipment, placement of the equipment, and methods have been reported previously)' 4 During biofeedback training the patient was instructed to increase intraabdominal pressure, to push as if defecating, and to inhibit contraction of the external anal sphincter in attempts to produce normal relaxations. Biofeedback training was stopped after relaxations of the external anal sphincter without visual feedback were accomplished in two successive biofeedback training sessions. Outcome evaluation. Outcome was evaluated 12 months after the start of treatment. The follow-up occurred during a clinic visit, by mailed questionnaire, or by telephone; the last month was rated. Nonresponders to the mailed questionnaire were interviewed by telephone. A structured questionnaire, eliciting information on the presence of soiling and the frequency and amount of soiling per week, the frequency and size of bowel movements in the toilet per week, the use of laxatives, the presence of abdominal pain, and daytime and nighttime urinary incontinence, was used and approved by the institutional human research review committee. The letter accompanying the questionnaire indicated that the data were obtained for research purposes. Patients were rated as recovered if they had three or more bowel movements/week and two or fewer soiling episodes
338
Loening-Baucke
The Journal of Pediatrics March 1996
Table I. Initial patient characteristics
Soilings/wk* Stool frequency/wk* Primary encopresis (%) Stool withholding(%) Abdominalpain (%) Urinary incontinence Daytime (%) Nighttime (%) Rectal impaction(%) Severe constipation(%)
T a b l e II. Twelve-month outcome
Def+ (n = 65)
Def(n = 74)
14 _+ 15 4+4 46 54 57
17 _+ 19 4- 5 39 59 51
37? 42t 94 31
19 23 99 55?
Def+, Patientsableto defecatethe rectalballoon;Def-, patientsunableto defecate the rectal balloon. *Values are expressedas mean-+ SD. "~p<0.03.
per month while laxative use was discontinued for at least 1 month. Patients were counted as not recovered if they had fewer than three bowel movements per week or were soiling more than twice a month or still needed laxatives. Patients were rated as successfully treated if they had three or more bowel movements a week, two or fewer soiling episodes a month, and no abdominal pain, irrespective of laxative use. Statistical analysis. For statistical analysis, we used the Wilcoxon rank sum test and signed rank test, the Fisher Exact Test, and logistic regression, with significance accepted at the 5% level. Results were expressed as mean +- SD. RESULTS Studies of balloon defecation. All 20 healthy children tested were able to defecate the rectal balloon. Sixty-five children with functional constipation and encopresis (47%) were able to defecate the balloon within 5 minutes, and 74 (53%) were unable to do so. Patients able and those unable to defecate the rectal balloon exerted significantly higher pressures on the rectal balloon (78 +_ 30 mm Hg) and strained significantly longer (9 +_ 4 seconds) than healthy children (43 +- 26 mm Hg; 4 + 1 seconds) (p <0.001). Patients unable to defecate the balloon exerted significantly higher pressures on the water-filled rectal balloon (84 _+ 30 mm Hg) than those able to defecate the balloon (71 -+ 28 mm Hg; p <0.03) and strained significantly longer (11 -+ 4 versus 8 +- 5 seconds; p <0.001), indicating that the inability to defecate the balloon was not due to weak straining efforts. Ninety-three percent of patients able to and 80% of those unable to defecate the balloon reported that they felt the presence of the 100 ml water-filled balloon (p <0.04), and 73% versus 39% reported that this feeling was an urge to defecate (p <0.001). Initial patient characteristics (Table I). Patients were grouped by their ability or inability to defecate the balloon.
Recovery (%) Soiling (%) Soiling episodes/wk* Constipation(%) Stool frequency/wk* On laxatives (%) Successful treatment(%)
Def+ (n = 65)
Def(n = 74)
p
51 31 1 _+2 18 7 _+4 34 57
34 43 2 +_3 34 6 _+4 43
0.03 0.16 0.09 0.054 0.08 0.29
39
0.04
Def+, Patientsableto defecatethe rectal balloon;Def-, patientsunableto defecate the rectal balloon. *Values are expressedas mean+- SD.
Soiling episodes per week and stool frequency per week were similar in both patient groups (p >0.8). Primary encopresis (defined as never having been reliably toilet trained), stool withholding, abdominal pain, and fecal impaction of the rectum were similarly distributed in both patient groups (p >0.1). Daytime and nighttime urinary incontinence were more frequently present in patients who could defecate the balloon (p <0.03). Severe constipation, defined by the presence of a palpable abdominal mass on initial examination, was more common in patients unable to defecate the balloon (55%), in comparison with patients who were able to defecate the balloon (31%) (p <0.004). Twelve-month outcome (Table II). Significantly more patients recovered if they were able than if they were unable to defecate the rectal balloon (51% vs 34%; p <0.03). Soiling, constipation, and laxative use were similar at follow-up in patients who could and could not defecate the balloon (p >0.05). The soiling frequency per week had decreased significantly and stool frequency per week had increased significantly at follow-up in both patient groups (p <0.001). Soiling frequency per week and stool frequency per week were not significantly different between the two patient groups (p >0.08). Successful treatment was also significantly more commonly observed in patients able (57%) than in those unable (39%) to defecate the rectal balloon (p <0.04). At 12-months of follow-up, significantly fewer patients complained of abdominal pain--15% of those able and 11% of those unable to defecate the balloon (p >0.6). In addition, daytime urinary incontinence had resolved significantly in both groups; 6% and 5%, respectively, still had some daytime urinary incontinence. Nighttime urinary incontinence resolved significantly in both groups (p <0.03) but was still more commonly present in patients able (23%) than in those unable (8%) to defecate the balloon (p <0.02). Univariate logistic regression revealed that the ability to defecate the 100 ml water-filled balloon (p <0.01) and having secondary encopresis (p <0.04) were significantlyrelated to recovery. Logistic regression revealed that the ability to defecate the balloon was significantly related to recovery (p
The Journal of Pediatrics Volume 128, Number 3
Loening-Baucke
339
Table Ill.Anorectal manometric function results
Threshold volumes (ml) Transientrectal sensation Initial urge Strong urge Constant relaxation Rectal contractility Anismus (%)
Control subjects (n = 32)
Def+ (n = 65)
Def(n = 74)
27 _+ 13 63 _+35~ 108 -+42 113 _+54 64 +_41 05
42 _+24* 95 -+ 64* 163 + 75* 178 _+76* t00 -+ 78* 49*
58 _+ 16",~ 122 _+74*,? 223 _+99",? 234 _+ 102",? 146 _+98*,? 78',t
Threshold volumesare expressedas mean-+ SD. Def+, Patientsable to defecate the rectal balloon;Def-, patientsunableto defecate the rectal balloon. *p <0.02comparedwith controlsubjects. t"p <0.03 comparedwith Def+. Sn= 16. <0.04; odds ratio 2.13, 95% confidence limits 1.06 and 4.29). When initial patient characteristics were entered into the model, only secondary encopresis, defined as encopresis occurring after the patient has been completely toilet trained, was significantly related to recovery (p <0.04; odds ratio 2.09, 95% confidence limits 1.04 and 4.23). No other patient characteristic met the 0.15 significance level for entry into the model. The ability of the balloon defecation test to predict recovery in children with constipation and encopresis had a sensitivity of 57%, a specificity of 60%, a positive predictive value of 0.51, and a negative predictive value of 0.66. Anorectalfunctions (Table III). The anorectal functions in both patient groups were significantly impaired, in comparison with those in healthy control subjects (p <0.02). The distention volumes were all significantly higher in patients who could not defecate the balloon than in those able to defecate the balloon: to produce a transient rectal sensation of balloon distention (p <0.03), an initial urge to defecate (p <0.001), a strong urge to defecate (p <0.001), and a constant relaxation of the external and internal anal sphincters (p <0.001), and to induce rectal contractility (p <0.002). Anismus, an abnormal contraction of the external anal sphincter and pelvic floor during straining for defecation, was significantly more common in patients unable to defecate the balloon than in those able to defecate the balloon (p <0.001). Univariate logistic regression revealed no significant relationship to recovery for the threshold volumes for initial urge and strong urge to defecate (p >0.06), the threshold volume for constant relaxation (p >0.1), the threshold volume to induce rectal contractility (p >0.1), the presence of anismus (p >0.09), or the presence of severe constipation (p >0.1). After the ability to defecate the balloon was entered into the regression model, the logistic regression revealed that no additional anorectal function measurement met the 0.15 significance level for entry into the model; these included the threshold volumes for strong urge to defecate (p >0.26), for constant relaxation (p >0.42), for the induction
of rectal contractility (p >0.36), and for the presence of anismus (p >0.59).
The effect of biofeedback treatment on 12-month outcome. We evaluated the outcome of biofeedback treatment in patients with anismus. Thirteen patients with anismus and able to defecate the balloon received biofeedback training; 11 of them learned to relax the external anal sphincter during straining for defecation, and seven of those recovered. Thirty-five patients with anismus and unable to defecate the balloon received biofeedback training; 28 of them learned to relax the external anal sphincter during straining for defecation, and nine of those recovered. Twenty patients who were able and 24 patients who were unable to defecate the balloon had anismus but did not receive biofeedback training; 40% and 38% recovered, respectively. Therefore the recovery rates were independent of biofeedback training in patients able and those unable to defecate the balloon (p >0.1). DISCUSSION We found that children with functional constipation and encopresis who were able to defecate the rectal balloon were twice as likely to recover 12 months after the start of treatment. However, the ability of the balloon defecation test to predict recovery in children with constipation and encopresis was little better than chance. Patients who had secondary encopresis were twice as likely to recover 12 months after the start of treatment. No other patient characteristic or anorectal function measurement was significantly related to recovery once the results of the balloon defecation test were entered into the regression model. There was no effect of biofeedback treatment on the 12-month recovery rates. We cannot explain the reason that urinary incontinence was more frequently present in patients able to defecate the balloon (p <0.03). Some of our previous studies7, 8 showed that severe constipation was a risk factor for persistence of constipation, encopresis, or both. This finding was not confirmed in one previous study4 or in the present study. The anorectal manometric data revealed significantly
340
Loening-Baucke
more impairment in anorectal functions in patients who could not defecate the balloon, which suggests that these impaire d anorectal functions were responsible for the inability to defecate the balloon. Anismus during defecation has been proposed as an important component of anorectal dysfunction in children with functional constipation and encopresis. Biofeedback treatment to reverse anismus to normal defecation dynamics had been heralded as the new treatment for children with functional constipation. 3, 17-19 However, learning to relax the external anal sphincter with biofeedback training did not improve the 12-month recovery rate over conventional treatment alone in patients with constipation, encopresis, and anismus in our current study. In a randomized, controlled study, we had previously found benefits of biofeedback treatment in 20 children with constipation, encopresis, and anismus. 3 In retrospect, the significant effect of biofeedback treatment on recovery in our initial study 3 might have been due to the excellent patienttherapist relationship, because a different biofeedback therapist in our laboratory could not accomplish similar results.4, 2o The long-term recovery rate (mean, 6 years) of these 20 biofeedback-treated patients was 52%, similar to the 62% recovery rate of patients with anismus given conventional treatment alone. 2° The failure of biofeedback treatment significantly to improve recovery rates confirms three controlled studies. 21-23 Treatment of functional constipation and encopresis requires teamwork and close follow-up, including encouragement, education, and fine-tuning of the conventional treatment program. Most patients who comply with a conventional treatment program will have improvement. Relapse after discontinuation of treatment is frequent. The rectal balloon defecation test is an easily performed, low-cost test to evaluate the coordinated act of defecation. Using the rectal balloon defecation test in children with functional constipation and encopresis, we found that patients who were able to defecate the rectal balloon were twice as fikely to recover 12 months after the start of treatment. Even though there was a significant difference in the recovery rates between patients who could and those who could not defecate the balloon, calculating predictive values showed that the balloon defecation test could not reliably predict recovery. REFERENCES
1. Levine MD. Children with encopresis: a descriptive analysis. Pediatrics 1975;56:412-6. 2. Taitz LS, Water JKH, Urwin OM, Molnar D. Factors associated with outcome in management of defecation disorders. Arch Dis Child 1986;61:472-7. 3. Loening-Baucke V. Modulation of abnormal defecation dynamics by biofeedback treatment in chronically constipated children with encopresis. J PEDL~TR1990;116:214-22.
The Journal of Pediatrics March 1996
4. Loening-Baucke V. Persistence of chronic constipation in children after biofeedback treatment. Dig Dis Sci 1991 ;36:15360. 5. Wald A, Chandra R, Chiponis D, Gabel S. Anorectal function and continence mechanisms in childhood encopresis. J Pediatr Gastroenterol Nutr 1986;5:346-51. 6. Emery Y, Descos L, Meunier P, Louis D, Valancogne G, Weil G. Constipation terminale par asynchronisme abdomino-pelvien; analyse des donndes 6tiologiques, cliniques, manomdtriques, et des r6sultats thdrapeutiques apr6s rddducation par biofeedback. Gastroenterol Clin BiN 1988;12:6-11. 7. Loening-Baucke V. Factors responsible for persistence of childhood constipation. J Pediatr Gastroenterol Nutr 1987;6: 915-22. 8. Loening-Baucke V. Factors determining outcome in children with chronic constipation and faecal soiling. Gut 1989;30:9991006. 9. Meunier P, Louis D, Jaubert de Beaujeu M. Physiologic investigation of primary chronic constipation in children: comparison with the barium enema study. Gastroenterology 1984;87: 1351-7. 10. Meunier P, Marechal JM, Jaubert de Beaujeu M. Rectoanal pressures and rectal sensitivity studies in chronic childhood constipation. Gastroenterology 1979;77:330-6. 11. Loening-Baucke V. Sensitivity of the sigmoid colon and rectum in children treated for chronic constipation. J Pediatr Gastroenterol Nutr 1984;3:454-9. 12. Loening-Baucke V. Abnormal rectoanal function in children recovered from chronic constipation and encopresis. Gastroenterology 1984;87:1299-304. 13. Loening-Baucke V, Cmikshank B. Abnormal defecation dynamics in chronically constipated children with encopresis. J PEDIATR1986;108:562-6. 14. Preston DM, Lennard-Jones JE. Anismus in chronic constipation. Dig Dis Sci 1985;30:413-8. 15. Barnes PRH, Lennard-Jones JE. Balloon expulsion from the rectum in constipation of different types. Gut 1985;26:1049-52. 16. Loening-Baucke V. Chronic constipation in children. Gastroenterology 1993:105:1557-64. 17. Louis D, Valancogne G, Loras O, Meunier P. Techniques et indications du biofeedback dans les constipations chez l'enfant. Psychol Med 1985;17:1625-7. 18. Steffen R. Schroeder TK. Paradoxical pnborectalis contraction in children. Dis Colon Rectum 1992;35:1193. 19. Keren S, Wagner Y, Heldenberg D, Golan M. Studies of manometric abnormalities of the rectoanal region during defecation in constipated and soiling children: modification through biofeedback therapy. Am J Gastroenterol 1988;83:827-31. 20. Loening-Baucke V. Biofeedback treatment for chronic constipation and encopresis in childhood: long-term outcome. Pediatrics 1995;96:105-10. 21. Wald A, Chandra R, Gabel S, Chiponis D. Evaluation of biofeedback in childhood encopresis. J Pediatr Gastroenterol Nutr 1987;6:554-8. 22. Nolan T, Catto-Smith A, Coffey C, Wells J. EMG biofeedback training in anismus-related encopresis does not produce sustained continence [Abstract]. Arch Pediatr Adolesc Med 1995; 149;48. 23. Cox DJ, Sutphen J, Borowitz S, Dickens MN, Singles J, Whitehead WE. Simple electromyographic biofeedback treatment for chronic pediatric constipation/encopresis: preliminary report. Biofeedback Self Regul 1994;19:41-50.
Objective: To evaluate whether the ability to defecate a rectal balloon might predict 12-month recovery in children with functional constipation and encopresis. Methods: We evaluated the ability to defecate within 5 minutes a 100 ml waterfilled rectal balloon by 20 healthy children and 139 children with functional constipation and encopresis. Results: All healthy children and only 47% of the patients were able to defecate the balloon. Twelve months after the start of treatment, 5 I% of patients able to and 34% of patients unable to defecate the balloon had recovered (p <0.03). Logistic regression revealed that the ability to defecate the balloon and a history of secondary encopresis were related to recovery (p <0.04). Patients who were unable to defecate the balloon or who did not recover had significantly more impairment in anorectal functions than those who were able to defecate the balloon or who did recover. The ability of the balloon defecation test to predict recovery had a sensitivity of 57%, a specificity of 60%, a positive predictive value of 0.51, and a negative predictive value of 0.66. Conclusion: Children with functional constipation and encopresis who were able to defecate the rectal balloon were twice as likely to recover. Even though there was a clinically significant difference in the recovery rates between patients who could and those who could not defecate the balloon, calculation of predictive values showed that the balloon defecation test could not reliably predict recovery. (J PEDIATR1996; 128:336-40)
Constipation with or without encopresis (any amount of fecal soiling in children with stool retention) represents a common problem in children, accounting for 3% of visits to the University of Iowa Hospitals and Clinics Pediatric OutPresented in part at the 95th Annual Meeting of the American Gastroenterological Association, May 15, 1995, and published in abstract form (Gastroenterology 1995;108:640A). Supported by grant No. M01-RR-00059 from the General Clinical Research Center Program, Division of Research Resources, National Institutes of Health, and the Children's Miracle Network Telethon. Reprint requests: Vera Loening-Baucke, MD, University of Iowa Hospitals and Clinics, Department of Pediatrics, JCP 2555, 200 Hawkins Dr., Iowa City, IA 52242. Copyright © 1996 by Mosby-Year Book, Inc. 0022-3476/96/$5.00 + 0 9•20•69851
336
patient Clinic and to a large general pediatric clinic in Boston, ~ and for 25% of visits to a pediatric gastroenterology clinic.2 Constipation in children can be defined by a stool frequency of less than 3 per week; constipation can also be defined by the presence of a large rectal stool mass (stool retention) with or without encopresis, even when the stool frequency is three or more per week. The ability to defecate depends on several factors, including the simultaneous actions of the voluntary external and the involuntary internal anal sphincters, the tonicity of the rectum, and the condition of the spinal cord. Some children with functional constipation and encopresis have ineffectual straining, and up to 50% contract rather than relax the external anal sphincter and pelvic floor during straining for defecation (termed anismus, abnormal defecation dynamics or patterns, rectoanal or rectosphincteric dyssynergia, and
The Journal o[ Pediatrics Volume 128, Number 3
spastic pelvic floor syndrome3-8). Others have a megarecturn, rectal atony, or both, which hinder the propulsion of stool from the rectum by peristalsis.9-12 These abnormalities impede defecation. A method to test the coordinated act of defecation is the rectal balloon defecation test. 1315 The aim of our study was to evaluate whether the ability to defecate, as demonstrated by the defecation of a 100 ml water-filled rectal balloon, might predict 12-month recovery in children with functional constipation and encopresis.
METHODS Study patients. Two hundred thirty-two consecutive patients (176 boys and 56 girls, 5 to 18 years of age, mean age 9 + 3 years) with functional constipation and encopresis (one or more soiling episodes per week) were examined in the University of Iowa Encopresis Clinic with the balloon defecation test between 1985 and 1993. One hundred thirtynine of them (103 boys and 36 girls, mean age 10 _+ 3 years) were reexamined 12 months after the start of treatment and constitute the study group. These patients were healthy except for problems related to constipation and encopresis. Children with Hirschsprung disease, hypothyroidism, mental deficiency, chronic debilitating diseases, or neurologic abnormalities, and children who had previous surgery of the colon were excluded. Thirty-two healthy children (16 boys and 16 girls, 6 to 16 years of age, mean age 11 + 3 years) served as control subjects. All were studied with anorectal manometry, and 20 of them underwent the balloon defecation test.
Anorectal studies Studies of balloon defecation. To simulate defecation of a stool from the rectum, we asked children to defecate a 100 ml water-filled rectal balloon while sitting on a toilet chaff', allowing 5 minutes for defecation. The rectal balloon was made from a latex party balloon (Latex Occidental, distributed by American Imports, Inc., San Antonio, Tex.) and was 2.5 x 3 cm when deflated and 5 x 5 cm in size when filled with 100 ml of water. The empty balloon was lubricated and inserted into an empty rectum. AnorectaI manometry. In addition to the balloon defecation test, anorectal functions were assessed with anorectal manometry in all children. The methods were reported previously.lll 3 We determined the threshold volumes (measured in milliliters of air) required to elicit a transient sensation of rectal balloon distention, an initial urge to defecate, a strong urge to defecate (critical volume), a sustained complete relaxation of the internal and external anal sphincters (constant relaxation), and rectal contractility (10 mm Hg or more, recorded 5 cm below the base of the rectal balloon) by stepwise addition of 10 ml air increments, up to 60 ml, and then 30 ml increments, each 10 to 15 seconds into the rectal bal-
Loening-Baucke
337
loon.3, 4, 11 When the critical volume was reached, but constant relaxation or rectal contractility had not yet occurred, the next higher volume was used as the volume of constant relaxation or rectal contractility. While lying in the left lateral position, the child was asked to bear down five times as if defecating and to squeeze (tighten up) five times in random order. If no increase in intraabdominal pressure (rectal pressure) during the defecation trials was recorded, the child was instructed to increase intraabdominal pressure. We defined defecation dynamics as normal if the integrated electromyogram and the anal pressure decreased during increased intraabdominal pressure in two or more of five defecation trials. Anismus was defined by the presence of a persistent increase in the integrated electromyogram and the anal pressure during bearing down in four or more of five defecation trials,y' 8, 13 Treatment. All patients received conventional treatment for constipation and encopresis similar to that used for the past 15 years in our encopresis clinic. 16 The treatment included disimpaction, education to relieve the social and psychologic problems faced by these children and their parents, prevention of reaccumulation of fecal material in the colon with the help of daily laxative use and an increase in dietary fiber, and scheduled toilet sittings. 16 Parents and children were instructed to keep a diary of bowel movements, fecal soiling, and medication used. Forty-eight patients with anismus received, in addition to conventional treatment, at least two and up to six biofeedback training sessions (mean 3 + 2). The equipment, placement of the equipment, and methods have been reported previously)' 4 During biofeedback training the patient was instructed to increase intraabdominal pressure, to push as if defecating, and to inhibit contraction of the external anal sphincter in attempts to produce normal relaxations. Biofeedback training was stopped after relaxations of the external anal sphincter without visual feedback were accomplished in two successive biofeedback training sessions. Outcome evaluation. Outcome was evaluated 12 months after the start of treatment. The follow-up occurred during a clinic visit, by mailed questionnaire, or by telephone; the last month was rated. Nonresponders to the mailed questionnaire were interviewed by telephone. A structured questionnaire, eliciting information on the presence of soiling and the frequency and amount of soiling per week, the frequency and size of bowel movements in the toilet per week, the use of laxatives, the presence of abdominal pain, and daytime and nighttime urinary incontinence, was used and approved by the institutional human research review committee. The letter accompanying the questionnaire indicated that the data were obtained for research purposes. Patients were rated as recovered if they had three or more bowel movements/week and two or fewer soiling episodes
338
Loening-Baucke
The Journal of Pediatrics March 1996
Table I. Initial patient characteristics
Soilings/wk* Stool frequency/wk* Primary encopresis (%) Stool withholding(%) Abdominalpain (%) Urinary incontinence Daytime (%) Nighttime (%) Rectal impaction(%) Severe constipation(%)
T a b l e II. Twelve-month outcome
Def+ (n = 65)
Def(n = 74)
14 _+ 15 4+4 46 54 57
17 _+ 19 4- 5 39 59 51
37? 42t 94 31
19 23 99 55?
Def+, Patientsableto defecatethe rectalballoon;Def-, patientsunableto defecate the rectal balloon. *Values are expressedas mean-+ SD. "~p<0.03.
per month while laxative use was discontinued for at least 1 month. Patients were counted as not recovered if they had fewer than three bowel movements per week or were soiling more than twice a month or still needed laxatives. Patients were rated as successfully treated if they had three or more bowel movements a week, two or fewer soiling episodes a month, and no abdominal pain, irrespective of laxative use. Statistical analysis. For statistical analysis, we used the Wilcoxon rank sum test and signed rank test, the Fisher Exact Test, and logistic regression, with significance accepted at the 5% level. Results were expressed as mean +- SD. RESULTS Studies of balloon defecation. All 20 healthy children tested were able to defecate the rectal balloon. Sixty-five children with functional constipation and encopresis (47%) were able to defecate the balloon within 5 minutes, and 74 (53%) were unable to do so. Patients able and those unable to defecate the rectal balloon exerted significantly higher pressures on the rectal balloon (78 +_ 30 mm Hg) and strained significantly longer (9 +_ 4 seconds) than healthy children (43 +- 26 mm Hg; 4 + 1 seconds) (p <0.001). Patients unable to defecate the balloon exerted significantly higher pressures on the water-filled rectal balloon (84 _+ 30 mm Hg) than those able to defecate the balloon (71 -+ 28 mm Hg; p <0.03) and strained significantly longer (11 -+ 4 versus 8 +- 5 seconds; p <0.001), indicating that the inability to defecate the balloon was not due to weak straining efforts. Ninety-three percent of patients able to and 80% of those unable to defecate the balloon reported that they felt the presence of the 100 ml water-filled balloon (p <0.04), and 73% versus 39% reported that this feeling was an urge to defecate (p <0.001). Initial patient characteristics (Table I). Patients were grouped by their ability or inability to defecate the balloon.
Recovery (%) Soiling (%) Soiling episodes/wk* Constipation(%) Stool frequency/wk* On laxatives (%) Successful treatment(%)
Def+ (n = 65)
Def(n = 74)
p
51 31 1 _+2 18 7 _+4 34 57
34 43 2 +_3 34 6 _+4 43
0.03 0.16 0.09 0.054 0.08 0.29
39
0.04
Def+, Patientsableto defecatethe rectal balloon;Def-, patientsunableto defecate the rectal balloon. *Values are expressedas mean+- SD.
Soiling episodes per week and stool frequency per week were similar in both patient groups (p >0.8). Primary encopresis (defined as never having been reliably toilet trained), stool withholding, abdominal pain, and fecal impaction of the rectum were similarly distributed in both patient groups (p >0.1). Daytime and nighttime urinary incontinence were more frequently present in patients who could defecate the balloon (p <0.03). Severe constipation, defined by the presence of a palpable abdominal mass on initial examination, was more common in patients unable to defecate the balloon (55%), in comparison with patients who were able to defecate the balloon (31%) (p <0.004). Twelve-month outcome (Table II). Significantly more patients recovered if they were able than if they were unable to defecate the rectal balloon (51% vs 34%; p <0.03). Soiling, constipation, and laxative use were similar at follow-up in patients who could and could not defecate the balloon (p >0.05). The soiling frequency per week had decreased significantly and stool frequency per week had increased significantly at follow-up in both patient groups (p <0.001). Soiling frequency per week and stool frequency per week were not significantly different between the two patient groups (p >0.08). Successful treatment was also significantly more commonly observed in patients able (57%) than in those unable (39%) to defecate the rectal balloon (p <0.04). At 12-months of follow-up, significantly fewer patients complained of abdominal pain--15% of those able and 11% of those unable to defecate the balloon (p >0.6). In addition, daytime urinary incontinence had resolved significantly in both groups; 6% and 5%, respectively, still had some daytime urinary incontinence. Nighttime urinary incontinence resolved significantly in both groups (p <0.03) but was still more commonly present in patients able (23%) than in those unable (8%) to defecate the balloon (p <0.02). Univariate logistic regression revealed that the ability to defecate the 100 ml water-filled balloon (p <0.01) and having secondary encopresis (p <0.04) were significantlyrelated to recovery. Logistic regression revealed that the ability to defecate the balloon was significantly related to recovery (p
The Journal of Pediatrics Volume 128, Number 3
Loening-Baucke
339
Table Ill.Anorectal manometric function results
Threshold volumes (ml) Transientrectal sensation Initial urge Strong urge Constant relaxation Rectal contractility Anismus (%)
Control subjects (n = 32)
Def+ (n = 65)
Def(n = 74)
27 _+ 13 63 _+35~ 108 -+42 113 _+54 64 +_41 05
42 _+24* 95 -+ 64* 163 + 75* 178 _+76* t00 -+ 78* 49*
58 _+ 16",~ 122 _+74*,? 223 _+99",? 234 _+ 102",? 146 _+98*,? 78',t
Threshold volumesare expressedas mean-+ SD. Def+, Patientsable to defecate the rectal balloon;Def-, patientsunableto defecate the rectal balloon. *p <0.02comparedwith controlsubjects. t"p <0.03 comparedwith Def+. Sn= 16. <0.04; odds ratio 2.13, 95% confidence limits 1.06 and 4.29). When initial patient characteristics were entered into the model, only secondary encopresis, defined as encopresis occurring after the patient has been completely toilet trained, was significantly related to recovery (p <0.04; odds ratio 2.09, 95% confidence limits 1.04 and 4.23). No other patient characteristic met the 0.15 significance level for entry into the model. The ability of the balloon defecation test to predict recovery in children with constipation and encopresis had a sensitivity of 57%, a specificity of 60%, a positive predictive value of 0.51, and a negative predictive value of 0.66. Anorectalfunctions (Table III). The anorectal functions in both patient groups were significantly impaired, in comparison with those in healthy control subjects (p <0.02). The distention volumes were all significantly higher in patients who could not defecate the balloon than in those able to defecate the balloon: to produce a transient rectal sensation of balloon distention (p <0.03), an initial urge to defecate (p <0.001), a strong urge to defecate (p <0.001), and a constant relaxation of the external and internal anal sphincters (p <0.001), and to induce rectal contractility (p <0.002). Anismus, an abnormal contraction of the external anal sphincter and pelvic floor during straining for defecation, was significantly more common in patients unable to defecate the balloon than in those able to defecate the balloon (p <0.001). Univariate logistic regression revealed no significant relationship to recovery for the threshold volumes for initial urge and strong urge to defecate (p >0.06), the threshold volume for constant relaxation (p >0.1), the threshold volume to induce rectal contractility (p >0.1), the presence of anismus (p >0.09), or the presence of severe constipation (p >0.1). After the ability to defecate the balloon was entered into the regression model, the logistic regression revealed that no additional anorectal function measurement met the 0.15 significance level for entry into the model; these included the threshold volumes for strong urge to defecate (p >0.26), for constant relaxation (p >0.42), for the induction
of rectal contractility (p >0.36), and for the presence of anismus (p >0.59).
The effect of biofeedback treatment on 12-month outcome. We evaluated the outcome of biofeedback treatment in patients with anismus. Thirteen patients with anismus and able to defecate the balloon received biofeedback training; 11 of them learned to relax the external anal sphincter during straining for defecation, and seven of those recovered. Thirty-five patients with anismus and unable to defecate the balloon received biofeedback training; 28 of them learned to relax the external anal sphincter during straining for defecation, and nine of those recovered. Twenty patients who were able and 24 patients who were unable to defecate the balloon had anismus but did not receive biofeedback training; 40% and 38% recovered, respectively. Therefore the recovery rates were independent of biofeedback training in patients able and those unable to defecate the balloon (p >0.1). DISCUSSION We found that children with functional constipation and encopresis who were able to defecate the rectal balloon were twice as likely to recover 12 months after the start of treatment. However, the ability of the balloon defecation test to predict recovery in children with constipation and encopresis was little better than chance. Patients who had secondary encopresis were twice as likely to recover 12 months after the start of treatment. No other patient characteristic or anorectal function measurement was significantly related to recovery once the results of the balloon defecation test were entered into the regression model. There was no effect of biofeedback treatment on the 12-month recovery rates. We cannot explain the reason that urinary incontinence was more frequently present in patients able to defecate the balloon (p <0.03). Some of our previous studies7, 8 showed that severe constipation was a risk factor for persistence of constipation, encopresis, or both. This finding was not confirmed in one previous study4 or in the present study. The anorectal manometric data revealed significantly
340
Loening-Baucke
more impairment in anorectal functions in patients who could not defecate the balloon, which suggests that these impaire d anorectal functions were responsible for the inability to defecate the balloon. Anismus during defecation has been proposed as an important component of anorectal dysfunction in children with functional constipation and encopresis. Biofeedback treatment to reverse anismus to normal defecation dynamics had been heralded as the new treatment for children with functional constipation. 3, 17-19 However, learning to relax the external anal sphincter with biofeedback training did not improve the 12-month recovery rate over conventional treatment alone in patients with constipation, encopresis, and anismus in our current study. In a randomized, controlled study, we had previously found benefits of biofeedback treatment in 20 children with constipation, encopresis, and anismus. 3 In retrospect, the significant effect of biofeedback treatment on recovery in our initial study 3 might have been due to the excellent patienttherapist relationship, because a different biofeedback therapist in our laboratory could not accomplish similar results.4, 2o The long-term recovery rate (mean, 6 years) of these 20 biofeedback-treated patients was 52%, similar to the 62% recovery rate of patients with anismus given conventional treatment alone. 2° The failure of biofeedback treatment significantly to improve recovery rates confirms three controlled studies. 21-23 Treatment of functional constipation and encopresis requires teamwork and close follow-up, including encouragement, education, and fine-tuning of the conventional treatment program. Most patients who comply with a conventional treatment program will have improvement. Relapse after discontinuation of treatment is frequent. The rectal balloon defecation test is an easily performed, low-cost test to evaluate the coordinated act of defecation. Using the rectal balloon defecation test in children with functional constipation and encopresis, we found that patients who were able to defecate the rectal balloon were twice as fikely to recover 12 months after the start of treatment. Even though there was a significant difference in the recovery rates between patients who could and those who could not defecate the balloon, calculating predictive values showed that the balloon defecation test could not reliably predict recovery. REFERENCES
1. Levine MD. Children with encopresis: a descriptive analysis. Pediatrics 1975;56:412-6. 2. Taitz LS, Water JKH, Urwin OM, Molnar D. Factors associated with outcome in management of defecation disorders. Arch Dis Child 1986;61:472-7. 3. Loening-Baucke V. Modulation of abnormal defecation dynamics by biofeedback treatment in chronically constipated children with encopresis. J PEDL~TR1990;116:214-22.
The Journal of Pediatrics March 1996
4. Loening-Baucke V. Persistence of chronic constipation in children after biofeedback treatment. Dig Dis Sci 1991 ;36:15360. 5. Wald A, Chandra R, Chiponis D, Gabel S. Anorectal function and continence mechanisms in childhood encopresis. J Pediatr Gastroenterol Nutr 1986;5:346-51. 6. Emery Y, Descos L, Meunier P, Louis D, Valancogne G, Weil G. Constipation terminale par asynchronisme abdomino-pelvien; analyse des donndes 6tiologiques, cliniques, manomdtriques, et des r6sultats thdrapeutiques apr6s rddducation par biofeedback. Gastroenterol Clin BiN 1988;12:6-11. 7. Loening-Baucke V. Factors responsible for persistence of childhood constipation. J Pediatr Gastroenterol Nutr 1987;6: 915-22. 8. Loening-Baucke V. Factors determining outcome in children with chronic constipation and faecal soiling. Gut 1989;30:9991006. 9. Meunier P, Louis D, Jaubert de Beaujeu M. Physiologic investigation of primary chronic constipation in children: comparison with the barium enema study. Gastroenterology 1984;87: 1351-7. 10. Meunier P, Marechal JM, Jaubert de Beaujeu M. Rectoanal pressures and rectal sensitivity studies in chronic childhood constipation. Gastroenterology 1979;77:330-6. 11. Loening-Baucke V. Sensitivity of the sigmoid colon and rectum in children treated for chronic constipation. J Pediatr Gastroenterol Nutr 1984;3:454-9. 12. Loening-Baucke V. Abnormal rectoanal function in children recovered from chronic constipation and encopresis. Gastroenterology 1984;87:1299-304. 13. Loening-Baucke V, Cmikshank B. Abnormal defecation dynamics in chronically constipated children with encopresis. J PEDIATR1986;108:562-6. 14. Preston DM, Lennard-Jones JE. Anismus in chronic constipation. Dig Dis Sci 1985;30:413-8. 15. Barnes PRH, Lennard-Jones JE. Balloon expulsion from the rectum in constipation of different types. Gut 1985;26:1049-52. 16. Loening-Baucke V. Chronic constipation in children. Gastroenterology 1993:105:1557-64. 17. Louis D, Valancogne G, Loras O, Meunier P. Techniques et indications du biofeedback dans les constipations chez l'enfant. Psychol Med 1985;17:1625-7. 18. Steffen R. Schroeder TK. Paradoxical pnborectalis contraction in children. Dis Colon Rectum 1992;35:1193. 19. Keren S, Wagner Y, Heldenberg D, Golan M. Studies of manometric abnormalities of the rectoanal region during defecation in constipated and soiling children: modification through biofeedback therapy. Am J Gastroenterol 1988;83:827-31. 20. Loening-Baucke V. Biofeedback treatment for chronic constipation and encopresis in childhood: long-term outcome. Pediatrics 1995;96:105-10. 21. Wald A, Chandra R, Gabel S, Chiponis D. Evaluation of biofeedback in childhood encopresis. J Pediatr Gastroenterol Nutr 1987;6:554-8. 22. Nolan T, Catto-Smith A, Coffey C, Wells J. EMG biofeedback training in anismus-related encopresis does not produce sustained continence [Abstract]. Arch Pediatr Adolesc Med 1995; 149;48. 23. Cox DJ, Sutphen J, Borowitz S, Dickens MN, Singles J, Whitehead WE. Simple electromyographic biofeedback treatment for chronic pediatric constipation/encopresis: preliminary report. Biofeedback Self Regul 1994;19:41-50.