Prognosis of constipation: clinical factors and colonic transit time
- 1Department of Paediatric Gastroenterology and Nutrition, Emma Children’s Hospital/Academic Medical Centre, Amsterdam, Netherlands
- 2Department of Clinical Epidemiology and Biostatistics, Academic Medical Centre, Amsterdam, Netherlands
- Correspondence to:
Dr F de Lorijn
Department of Paediatric Gastroenterology and Nutrition, Emma Children’s Hospital/Academic Medical Centre, Meibergdreef 9, 1105 AZ Amsterdam, Netherlands;
- Accepted 23 December 2003
Background: Measurement of colonic transit time (CTT) is sometimes used in the evaluation of patients with chronic constipation.
Aim: To investigate the relation between symptoms and CTT, and to assess the importance of symptoms and CTT in predicting outcome.
Methods: Between 1995 and 2000, 169 consecutive patients (median age 8.4 years, 65% boys) fulfilling the criteria for constipation were enrolled. During the intervention and follow up period, all kept a diary to record symptoms. CTT was measured at entry to the study.
Results: At entry, defecation frequency was lower in girls than in boys, while the frequency of encopresis episodes was higher in boys. CTT values were significantly higher in those with a low defecation frequency (⩽1/week) and a high frequency of encopresis (⩾2/day). However, 50% had CTT values within the normal range. Successful outcome occurred more often in those with a rectal impaction. CTT results <100 hours were not predictive of outcome. However, those with CTT >100 hours were less likely to have had a successful outcome.
Conclusion: The presence of a rectal impaction at presentation is associated with a better outcome at one year. A CTT >100 hours is associated with a poor outcome at one year.
- CTT, colonic transit time
- RSTT, rectosigmoid transit time
- FNRFS, functional non-retentive faecal soiling
Constipation is a common problem in children, accounting for about 3% of consultations in an average paediatric practice and as much as 25% in a paediatric gastroenterology clinic.1–3 No specific organic cause can be found in approximately 90% of the children.4,5 The diagnosis is mainly based on clinical history and physical examination. Patients and/or their parents refer to the number of stools per week, to stool volume, to difficulty in defecation, and/or to sensation of abdominal fullness.4 Several tests have been developed to objectify these complaints.6–10
A plain abdominal radiograph is frequently used to confirm the presence of retained stool or enlargement of the colon or rectum.7,8 However, the value of abdominal x ray examination in this setting is a matter of debate.11 Assessment of total and segmental colonic transit time (CTT) using radio-opaque markers, is a non-invasive method which provides information about colorectal motor function.6,9,10 Furthermore, this technique has been used to localise a delayed transit in the colon and to evaluate the response to treatment.9,11–13 Some investigators have shown a good relation between symptoms of constipation and CTT in adults.14,15 However, others report a significant difference between previous reported symptoms and the CTT results.9,16 No previous studies have investigated the prognostic value of CTT measurement.
Our study had two main objectives. First, to investigate the relation between symptoms and CTT. Second, to evaluate the possible relation between symptoms and CTT, and the outcome after one year of follow up.
PATIENTS AND METHODS
All patients referred to our paediatric gastrointestinal outpatient clinic with constipation between 1995 and 2000 were eligible. They were referred by family practitioners, paediatricians, psychiatrists, and school doctors. At entry, patients had at least two of the following: (1) defecation <3/week; (2) encopresis episodes >1/week; (3) passing of very large stool every 7–30 days; and (4) a palpable abdominal or rectal faecal mass.17 All were at least 5 years of age. Children with Hirschsprung’s disease, spinal and anal anomalies, previous colon surgery, metabolic or renal abnormalities, mental retardation, or using drugs other than laxatives were excluded. All patients and parents gave written informed consent. The medical ethics committee of the hospital approved the protocol.
Colonic transit time
At entry, all children underwent a CTT study using the method described by Metcalf and colleagues.10 Treatment with oral or rectal laxatives was discontinued for at least four days before the test; during this period they took one sachet of fibre (Volcolon, 6 g) each day. They then ingested a capsule with containing 20 radio-opaque markers on three consecutive mornings. Abdominal x ray examinations were performed on days 4 and 7 in the morning. Additional abdominal x ray examinations were performed on days 10, 13, and 16 if more than 20% of the markers remained on the previous film. x Ray localisation of markers was based on the identification of bony landmarks and gaseous outlines as described by Arhan and colleagues.6 Markers were counted in the right, left, and rectosigmoid regions, and mean segmental transit times were calculated according to a previously described formula.6,10 The normal ranges for total and segmental transit times were based on the upper limits (mean + 2 SD) from a study in healthy children.6 Based on this study, a CTT of more than 62 hours was considered delayed. The upper limits of the normal range for right colon, left colon, and rectosigmoid transit time were 18, 20, and 34 hours, respectively.6
Medical history and physical examination
One week prior to entry and throughout the study the child and parents kept a diary in which defecation frequency, encopresis frequency, consistency and size of stool, and pain during defecation were recorded. Associated symptoms such as abdominal pain, appetite suppression, absence of urge to defecate, and enuresis were also noted.
At entry, abdominal and rectal examinations were performed. Abdominal distension and palpable abdominal faecal masses were noted. Anal tone and the presence of a faecal impaction were recorded.
Treatment and follow up
The treatment regimens consisted of a high fibre diet, toilet advice (attempt defecation on the toilet for five minutes after meals), laxatives, and biofeedback training or anorectal manometry.17 Oral laxative therapy consisted of an initial dosage of 6 g lactulose per day, increased stepwise to a maximum of 24 g if necessary. Enemas were given if severe rectal faecal impaction was present. A reward system was also employed.
The treatment intervention lasted for eight weeks. A detailed medical history was obtained during the intervention period, and at 6 and 12 months after entry. When necessary, children were also seen at other times at the outpatient clinic.
Definition of success
Treatment was considered to be “successful” if patients had three or more bowel movements weekly and fewer than one encopresis episode in a two week period, having discontinued laxatives for at least one month.
Baseline characteristics of the cohort were analysed in a descriptive way. Median values and 25th and 75th or 10th and 90th centiles were used if the distribution of continuous variables was skewed. Non-parametric (Mann-Whitney U and Kruskal-Wallis) and χ2 statistics were used to test for differences between groups. Logistic regression models were used to examine possible factors associated with success at one year. The following factors were considered: defecation frequency (in three groups: ⩽ 1/week, 1–3/week, and ⩾3/week), encopresis frequency (in four groups: no encopresis, <1/day, 1–2/day, ⩾2/day), the presence of night-time encopresis, the production of large stools, the presence of an abdominal or rectal mass, and CTT >100 hours. Odds ratios with 95% confidence intervals were used to express the strength of the associations.
A total of 169 consecutive children (65% boys) were enrolled. Table 1 presents their baseline characteristics. The median age at intake was 8.4 years. The median defecation frequency at intake was lower in girls than in boys (1.0 v 2.0 times per week; p = 0.03), while an encopresis frequency of more than twice weekly was reported more often in boys (94% v 73%; p = 0.0002). Of the total group, 65% reported the passage of large stools every 7–30 days. Night-time encopresis occurred in 37%. A rectal impaction was present in 32%.
Table 2 shows the total and segmental colonic transit times at entry to the study. Approximately 50% had a total CTT in the normal range. No significant differences were found between boys and girls in the CTT or rectosigmoid transit time (RSTT). In 86% of those with an abnormal CTT, the RSTT was also prolonged. In children with a normal CTT, a delay in RSTT was present in only 14%. There was a significant correlation between CTT and RSTT (Pearson correlation coefficient 0.88, p = 0.0002).
In the total group, a CTT >100 hours was found in 22%. Of these, 92% had a delayed RSTT and 39% had a delay in all segments.
Correlation between CTT and symptom severity
At the first visit, children with a very low defecation frequency (⩽1/week) showed significantly prolonged median CTT (74 hours) compared to children with a defecation frequency of 1–3/week (50 hours) and a defecation frequency ⩾3/week (49 hours) (p = 0.001). In addition, in patients with ⩾2 episodes of encopresis per day, median CTT values were significantly delayed (70 hours) compared to the group with 1–2 episodes of encopresis per day (50 hours), the group with <1 episode of encopresis per day (52 hours), and those without encopresis (49 hours) (p = 0.003) (table 3). Furthermore, a significantly higher CTT was present in children with night-time encopresis and in those with a rectal mass (respectively p<0.0001 and p<0.0001) (table 3). As table 3 shows, similar associations were found between clinical symptoms and rectosigmoid transit time.
Clinical outcome and prognostic value
After one year 50% of the boys and 72% of the girls had a successful outcome (relative risk (RR) boys v girls 0.70; 95% CI 0.55 to 0.90). These percentages increased to 61% in boys and to 80% in girls (RR 0.77; 95% CI 0.63 to 0.93) when patients were included who had a normal defecation frequency without encopresis but where still using laxatives.
The independent effect of various prognostic factors was analysed in a multivariate model predicting the probability of a successful outcome after one year. Boys were less likely to be successfully treated at one year (OR 0.34; CI 0.16 to 0.70). The presence of a mass on rectal examination was a positive predictive sign for success (OR 3.39; CI 1.30 to 8.83). In those with a CTT <100 hours at entry, the transit time did not predict outcome at one year. However, those with a CTT >100 hours were less likely to have a successful outcome (OR 0.31; CI 0.12 to 0.85). Other possible prognostic factors such as a low defecation frequency, the presence of encopresis, and large stools at entry were not associated with outcome (table 4).
This study shows that more severe symptoms, such as a relatively low defecation frequency, or a high encopresis frequency and the presence of a palpable rectal mass strongly correlate with a prolonged CTT and RSTT. Gender (girls) and a palpable rectal mass at entry were associated with a successful outcome at one year. An abdominal x ray examination with the use of radio-opaque markers at intake had no predictive value. Measurement of CTT did not predict outcome if less than 100 hours. In contrast, a CTT above 100 hours predicted a poor outcome at one year.
The ingestion of radio-opaque markers and the use of one or more abdominal x ray examinations allow the differentiation between children with normal or prolonged CTT.18 Several techniques have been used to measure CTT in patients with gastrointestinal disorders and have proved to be reliable and reproducible.19 In our study we used the three day Metcalf marker method. x Ray examinations were repeated with a three day interval until more than 80% of the markers disappeared.10 With this method, the radiation exposure is relatively high in children with severe constipation. In future studies, we will use the Bouchoucha method in which the patient ingests one capsule with 10 markers on six consecutive days. Subsequently, only one abdominal x ray examination at day 7 is needed to calculate CTT.20
In this study CTT was measured without first cleansing the colon. A recent study in adults with constipation showed no difference in distribution patterns of the markers before and after cleansing of the colon.21
In agreement with previous studies in adults and children we found that severe symptoms of constipation strongly correlated with prolonged CTT and RSTT.12,15,22 It has been suggested that constipation in children is mostly caused by the conscious or unconscious postponement of defecation (withholding behaviour). This could be learned behaviour due to pain with evacuation of a large faecal bolus. Prolonged stool retention in the rectum might result in increased anorectal sampling with failure of contraction of the external anal sphincter and consequent soiling. It is likely that the associated delay in RSTT leads to secondary prolongation of CTT.
In accordance with earlier studies in children with constipation a normal CTT was found in approximately 50% of the patients.11,22–24 It has been suggested that adults complaining of constipation with normal transit times have a high incidence of psychosocial disturbance and may not be truly constipated.25 In children, however, van der Plas et al showed that reported symptoms correlate well with the actual bowel habit.26 In a recent study we showed that colonic transit patterns, including normal colonic transit were not indicative of abnormal behaviour.27 Normal colonic transit time in constipated children is probably due to the fact that the measurement, even though it represents bowel pattern over a one week period, still remains a snapshot observation. Furthermore, the upper time limit for normal transit, as suggested by Arhan et al, is based on only 23 healthy children, and so may be misleading.6 Papadopoulou et al suggested that the markers themselves might accelerate transit, causing a false negative result.13
In our experience this marker test has proved useful in differentiating retentive from non-retentive defecation disorders, such as constipation and functional non-retentive faecal soiling.28 A normal CTT in combination with a normal defecation frequency and no rectal mass indicates the latter diagnosis.22 Such children are best treated with a toilet training programme without laxatives.29 Furthermore, treatment of CTT is useful in cases in which information is unreliable, such as in eating disorders.
After one year, 58% of the patients in our study had been treated successfully. This was in accordance with success rates in other long term follow-up studies.30–32 The overall success rate was 68% when patients without symptoms of constipation were included. In a recent long term study, more than 50% of the children with constipation experienced at least one relapse within the five years of first treatment.33
To our surprise, the presence of a rectal mass at presentation was a positive predictor of success. After removal of the mass most children experience less pain with defecation. It is likely that rectal control of sensation and defecation improves after removal, and there is a decrease in encopresis and abdominal pain.
A CTT longer than 100 hours predicted a poor outcome at one year. Almost 40% of these children had a delay in transit in all segments. Some patients, mainly adolescent girls, might suffer from idiopathic slow transit constipation.34–36 This severe form of constipation presents with a gradual reduction in bowel frequency and increasing abdominal pain.
In conclusion, the diagnostic and prognostic role of CTT measurements is limited. Only those with a CTT above 100 hours have a predictably poor outcome.