This issue of The Journal of Pediatric Gastroenterology and Nutrition contains three original articles on the evaluation and treatment of children with chronic retentive constipation. The article by Fishman et al. (1) is a retrospective review of data collected on a large group of children with constipation which attempts to tease out from an old database at The Boston Children's Hospital the behavioral and social antecedents of children with either primary or secondary encopresis. The article by Borowitz et al. (2) evaluates three approaches to constipation of escalating intensity—standard medical management, standard management plus toileting behavior therapy, and medical and behavioral therapy plus anal sphincter biofeedback training. The article by Loening-Baucke (3) compares the response of children with constipation two medical treatments—milk of magnesia or polyethylene glycol. The steady flow of articles dealing constipation reflects not only the frequency with which these patients come to our offices for care, but also the lack of a truly effective therapy.
Because the present medical therapy for constipation is incompletely effective (to put it kindly), Fishman et al. (1) asked whether these patients could be identified by a predictable set of characteristics that would allow prognostication at least and presymptomatic identification at best. These are certainly worthy goals. Because therapy is so poor, maybe early recognition and prevention is the direction we should take. The authors found that children with primary encopresis and children with secondary encopresis did not differ greatly in regard to premorbid characteristics; however, the group with primary encopresis had more difficult and interrupted toilet training, more problems with hard stools and straining, more abdominal pain, and fear of the toilet. Because another definition of “primary encopresis” could be “failure of toilet training,”, the study results strike me as fairly circular. It is only reasonable that a child who is unsuccessful in toilet training (i.e., has primary encopresis) should have had more trouble with and more interruption of toilet training. The other attributes of the children with primary encopresis do suggest that our primary care colleagues and we should identify and aggressively treat the healthy child whose mother reports grunting and straining because these data suggest but do not prove that such an approach may prevent later problems.
The two therapeutic studies are well-done, prospective comparisons of standard treatment programs. Although Borowitz et al. (2) have designed an original behavioral program that they will later be testing, behavior modification has been part of our therapies for many years. The finding of Borowitz et al. is that medical therapy seems to be more effective when given in combination with an intensive behavior modification program that includes education and supervised toileting. The finding of Loening-Baucke (3) is that therapy with polyethylene glycol, a relatively new product, is at least equal in effectiveness to therapy with milk of magnesia, an old standby. These two studies have been well designed and well conducted. They are prospective and randomized. The investigators were not blinded in either study. They have the authority of at least 12 months careful follow-up. The problem with these and all therapy studies is that in the end, the rate of “cure” in these children is only in the range of 50% to 60%. Studies performed many years ago report this kind of “cure rate” with other therapies, ranging from psychotherapy to enemas. Indeed, as noted by the sage of constipation, Dr. Melvin Levine back in 1975 (4), compliance with the regimen, whatever it may be, is the only independent predictor of response.
These recent articles discourage me a little. We have not made much therapeutic progress in the office management of simple, chronic retentive constipation despite numerous studies on diagnosis and therapy. No matter what we do, about 60% of our patients do well, and the other 40% have poorly controlled symptoms. If we are going to get beyond the 60% cure rate, it seems we should try something else in addition to modifications of the plumber's advice to “get them empty and keep them empty.” It is easy to suggest the right study to do next, and in practice difficult to accomplish. However, because constipation may affect up to 3% of children, perhaps we should work with our primary care colleagues to follow prospectively a cohort of 1,000 newborns and try to find out whether there are recognizable antecedents of this condition. For example, what if we checked 1,000 newborn infants for the threshold of rectal sensation? Would we possibly identify a group of infants with decreased rectal sensitivity who will later prove to be at increased risk for retentive constipation? There are thousands of children with this disorder. There must be a way to identify patients at risk and provide preventative therapy.
1. Fishman L, Rappaport L, Cousineau D, et al. Early constipation and toilet training in children with encopresis. J Pediatr Gastroenterol Nutr 2002; 34:385–88.
2. Borowitz S, Cox DJ, Sutphen JL, et al. Treatment of childhood encopresis: a randomized trial comparing three treatment protocols. J Pediatr Gastroenterol Nutr 2002; 34:378–84.
3. Loening-Bacuke V. Polyethylene glycol without electrolytes for children with constipation and encopresis. J Pediatr Gastroenterol Nutr 2002; 34:372–77
4. Levine MD. Children with encopresis: a descriptive analysis. Pediatrics 1975; 56:412–6.