Studies using radiopaque markers
A diagnostic case control study (2006) conducted in the Netherlands32 (2006) [EL=III] assessed the intra- and interobserver variability and the diagnostic accuracy of the Leech method of identifying children with functional constipation. The study included 89 consecutive children (median age 9.8 years) with the patients group comprising 52 constipated children. The 37 children in the control group fulfilled the criteria for functional abdominal pain (FAP) (n=6) and for functional non-retentive faecal incontinence (FNRFI) (n=31).
The Leech method to diagnose constipation in plain abdominal radiography was compared to the colonic transit time (CTT) with radiopaque markers. The mean Leech score (using the first score) was significantly higher in constipated children than in the control group (10.1 versus 8.5; P = 0.002). The mean CTT was significantly longer in constipated children than in the control group (92 hours versus 37 hours; P < 0.0001). The Leech method showed a sensitivity of 75% and a specificity of 59%. The positive predictive value and the negative predictive value were 72% and 63% respectively. The CTT showed a sensitivity of 79% and a specificity of 92% (cut off point 54 hours as per study). Using a cut off point of 62 hours (as per literature) the sensitivity decreased to 71% whereas the specificity improved to 95%. The positive predictive value was 69% and the negative predictive value was 97%. The area under the curve ROC was significantly smaller for the Leech method compared to the CTT (0.68, 95% CI 0.58 to 0.80 versus 0.90, 95% CI 0.83 to 0.96; P = 0.00015).
A diagnostic case control study conducted in China38 (2005) [EL=III] investigated the difference in CTT between constipated children and normal healthy controls to elicit its significance in assessing the dynamics of the whole gastrointestinal tract and each segment. The study included 96 children. There were 28 patients (gender not reported, mean age 6 years, age range 3 to 14) with confirmed functional constipation and 68 controls (38 boys, mean age 6 years, age range 3 to 13) with normal frequency and character of evacuation.
All children underwent CTT with radiopaque markers. No other tests or variables were used as a reference or comparator. Total CTT was significantly longer in patients compared to controls (mean 59.9 hours ± 2.3 versus 14.8 hours ± 0.8; P < 0.01). All segmental transit times were also significantly longer in patients compared to controls (right colon: mean 20.3 hours ± 1.2 versus 7.3 hours ± 1.1; P < 0.01); (left colon: mean 12.8 hours ± 1.7 versus 3.4 hours ± 0.8; P < 0.01); (rectosigmoid: mean 26.8 hours ± 1.4 versus 4.1 hours ± 1.2; P < 0.01).
A diagnostic prospective case series conducted in the Netherlands39 (2004) [EL=III] investigated the relation between symptoms of chronic constipation and CTT and evaluated the possible relation between symptoms and CTT and outcome after 1 year of follow up. The patients were 169 consecutive children (65% boys, median age 8.4 years) with chronic idiopathic constipation who underwent CTT. The following clinical variables were also recorded: defecation frequency, encopresis frequency, night-time encopresis and presence of a rectal mass on physical examination.
The total median CTT was 58 hours (25th to 75th centiles were 37 to 92). Forty-seven percent of the children had a delayed total CTT (more than 62 hours). Transit times for ascending colon, descending colon and rectosigmoid were 10 hours (5 to 16 hours), 10 hours (5 to 18 hours) and 32 hours (18 to 63 hours) respectively. Twenty-one percent of the children had delayed transit in the ascending colon (more than 18 hours), 22% in the descending colon (more than 20 hours) and 48% in the rectosigmoid (more than 34 hours). There were no significant differences in any of the outcomes between boys and girls. Children with a defecation frequency of 0 to1 per week (n=79) had a significantly longer CTT and rectosigmoid transit time (RSTT) compared to children with defecation frequencies of more than 1 to 3 times per week (n=55) and 3 or more times per week (n=35), (median CTT: 74 hours versus 50 hours and 49 hours; P = 0.001), (median RSTT: 38 hours versus 30 hours and 28 hours; P = 0.009). Children with an encopresis frequency (day and night) of 2 or more times per day (n=79) had significantly longer CTT and RSTT compared to children with an encopresis frequency of 1 to 2 times per day (n=48), children with an encopresis frequency of less than once per day (n=24) and children with no encopresis at all (n=18) (median CTT: 70 hours versus 50, 52 and 49 hours respectively; P = 0.003), (median RSTT: 38 hours versus 30, 31 and 24 hours respectively; P = 0.03).
Children with night time encopresis (n=63) had significantly longer CTT and RSTT compared with children without night time encopresis (n=106), (median CTT: 74 hours versus 47 hours; P < 0.0001), (median RSTT: 46 hours versus 28 hours; P < 0.0001). Children with a rectal mass present on physical examination (n=51) had significantly longer CTT and RSTT compared to children with no rectal mass (n=118), (median CTT: 86 hours versus 48 hours; P < 0.0001), (median RSTT: 64 hours versus 28 hours; P < 0.0001).
There were significant baseline differences between boys and girls. Median defecation frequency at intake was lower in girls than boys (1.0 versus 2.0 times per week; P = 0.03) and encopresis frequency more than twice weekly was reported more often in boys (94% versus 73%; P = 0.0002). More girls than boys reported no encopresis at all (20% versus 6%; P < 0.05).
A diagnostic case control study conducted in Brazil40 (2004) [EL=III] evaluated symptoms and clinical findings in a prospective series of adolescents with functional constipation and aimed to identify colonic disorders by measuring total and segmental colonic transit times with radiopaque markers. The study included 61 adolescents. Patients were 48 children (13 boys, mean age 14 years, range 12 to 18 years) with complaints of constipation for 1 year or longer. Controls were 13 children (9 boys, age not reported) with no digestive complaints who participated in a previous study by the same authors. All children underwent CTT with radiopaque markers and this was related to clinical variables.
Seventeen percent of the children were diagnosed with normal colonic transit, 60% with slow colonic transit, 13% with pelvic floor dysfunction and 10% with slow colonic transit and pelvic floor dysfunction. Total CTT (in hours) was significantly longer in constipated children compared to the healthy controls (mean 62.9 ± 12.6, median 69, range 62.9 to 12.6 versus mean 30.2 ± 13.2, median 27.5, range 10.8 to 50.4; P < 0.001). Segmental transit times (in hours) were also significantly longer in constipated children compared to the healthy controls for both the right and the left colon (right colon: mean 18.6 ± 15, median 13.2, range 12 to 54 versus mean 6.7 ± 3.9, median 4.8, range 1.2 to 12; P = 0.001); (left colon: mean 24.3 ± 13.7, median 22.8, range 2.4 to 51.6 versus mean 7.9 ± 7.8, median 7.2, range 0 to 28.8; P < 0.001).
There were no significant differences between constipated and non-constipated children for the rectosigmoid segment. The interval (in days) between evacuations was significantly longer for children with slow colonic transit compared to children with pelvic floor dysfunction (mean 7.7 ± 6.6 versus mean 3.7 ± 2.4; P < 0.003).
A faecal mass palpable at initial examination was statistically associated with slow colonic transit (P = 0.03). Other clinical variables were not statistically associated with a delay in either colon or rectosigmoid transit: onset of constipation, scybalous faeces, large volume, faecaloma, anal bleeding, soiling, previous use of laxative, suppositories or enemas, history of constipation in family, anal fissure, daily ingestion of fibre, sex, age and skin colour.
A diagnostic case control study conducted in Spain41 (2002) [EL=III] evaluated the use of a colonic motility study easily applied in daily clinical practice to more clearly define patients with this disorder. Sixty-eight children aged 2 to 14 years were included. Patients were 38 children with a history of chronic idiopathic constipation age more than 6 months, with or without secondary encopresis, refractory to conventional treatment. Controls were 30 children with normal bowel habits who underwent abdominal radiography as part of a clinical study with normal results. All children underwent CTT with radiopaque markers. No reference test was used but results were related to the frequency of defecation.
Patients had a significantly longer CTT (in hours) than controls (mean 49.57 ± 25.38, range 15.6 to 122.4 versus mean 29.08 ± 8.30, range 14.4 to 50; P < 0.001). Patients also had a significantly longer transit time (in hours) in both the left colon and the rectosigmoid compared to controls (left colon: mean 15.41 ± 13.13, range 2.4 to 32 versus mean 6.60 ± 6.20, range 2.4 to 24; P = 0.01); (rectosigmoid: mean 24.20 ± 16.77, range 4.8 to 69.6 versus mean 14.96 ± 8.70, range 2.4 to 19.2; P = 0.01). There were no significant differences in segmental transit time for the right colon between patients and controls.
Patients with a prolonged total CTT (n=19) were significantly younger at onset of constipation when compared to patients with a total CTT within reference values (n=19) (mean 1.77 years, SD 0.88 years versus mean 2.54 years, SD 1.18; P < 0.05). Significantly more patients with a prolonged total CTT (n=19) had a family history of constipation when compared to patients with a total CTT within reference values (n=19) (79% versus 21%; P < 0.01). An abdominal mass was found in significantly more patients with a prolonged total CTT (n=19) compared to patients with a total CTT within reference values (n=19) (93.8% versus 60%; P < 0.05). Encopresis was significantly more frequent in patients with a prolonged total CTT (n=19) compared to patients with a total CTT within reference values (n=19) (mean 0.60 episodes per night, SD 0.91 versus mean 0.10 episodes per night, SD 0.44; P < 0.05). No significant differences between patients and controls were found for age, age at diagnosis, gender, defecations per week, pain at defecation, enuresis, anal fissure, rectal mass or encopresis episodes per day, mean daily fibre intake and calorie consumption. A statistically significant inverse correlation was observed between total CTT and the number of weekly defecations (correlation coefficient r=0.68, P < 0.001). Two children from the patients group did not complete the study.
A diagnostic case control study conducted in Brazil42 (1998) [EL=III] measured total and segmental colonic transit time in constipated adolescents and compared the results with those in non-constipated children. Twenty-six adolescents aged 12 to 18 years were included in the study. Patients were 13 children with a history of constipation of at least one year of duration and controls were 13 children with no digestive complaints. There were nine boys in each group. All children underwent total and segmental CTT with radiopaque markers. Clinical variables were recorded.
The total CTT (in hours) was significantly longer in constipated children compared to non-constipated children (mean 58.25 ± 17.46, median 68.4, range 27.6 to 72 versus mean 30.18 ± 13.15, median 27.5, range 10.8 to 50.4; P < 0.001). Segmental transit times (in hours) for the right and left colon were also significantly longer in constipated children compared to non-constipated children (right colon: mean 15.97 ± 12.48, median 13.7, range 2.4 to 43.2 versus mean 6.74 ± 3.91, median 7.2, range 1.2 to 12; P = 0.03); (left colon: mean 24.74 ± 13.39, median 25.7, range 7.2 to 51.6 versus mean 7.94 ± 7.82, median 7.2, range 0 to 28.8; P < 0.001). There were no significant differences between the two groups for the transit time in rectosigmoid. The interval between stools was significantly longer for constipated children compared to non-constipated children (5.8 ± 2.3 days versus daily; P < 0.01). There were no significant differences between the two groups regarding: age, weight and height, bulky or small stools, encopresis, rectal mass, intense use of laxatives, bowel movements per week and mean daily intake of fibres.
A diagnostic case control study conducted in Poland43 (2007) [EL=III] determined whether a new method of ultrasound (US) assessment of stool retention could be used as a method of identifying children with functional chronic constipation and whether children with an enlarged rectum and colon (as seen on US) should be referred for further procedures such as proctoscopy and assessment of CTT. The study was conducted at a gastroenterology outpatient clinic and 225 children were enrolled, including 120 children (mean age 6.25 years) with chronic constipation who were compared to 105 children with a normal defecation pattern (mean age 8.25 years). Chronic constipation was diagnosed based on history and physical examination. In all patients the defecation disorders had persisted for longer than 6 months. All patients fulfilled the Rome II criteria for defecation disorders. The control group did not differ from the patients in gender but the comparison regarding age is not clearly reported.
Children underwent abdominal US. Children with a US diagnosis of megarectum, faecal impaction and enlarged colon were referred for proctoscopy and measurement of colonic transit time. Children with faecal impaction (as per US) had significantly longer average segmental transit time for the rectum, sigmoid and left colon (P < 0.001, P = 0.0015 and P = 0.0104 respectively). There was no statistically significant difference for the right side of the colon. Children with an overfilled splenic flexure on US had a significantly longer transit time in the left side of the colon (P = 0.0029).
A diagnostic case control study conducted in The Netherlands44 (1996) [EL=III] investigated the presence of slow colonic transit in children with constipation using radiopaque markers. The study included 148 children. Patients were 94 children (63 boys, mean age 8 years, range 5 to 14 years) with complaints of constipation with or without encopresis, encopresis alone or recurrent abdominal pain. Controls were 54 healthy children (10 boys, mean age 11 years, range 7 to 15 years). All children underwent CTT with radiopaque markers and their results were related to the presence of clinical symptoms.
Based on the CTT results 24 children were diagnosed with paediatric slow transit constipation (PSTC) and 70 children with normal delayed transit constipation (NDTC). The total CTT (in hours) was median 189 with a range of 104.4 to 380.4 for children with PSTC and median 46.8 with a range of 3.6 to 99.6 for children with NDTC (n=70). Median segmental transit time (in hours) in the right colon was 27.0 with a range of 3.6 to 60 for children with PSTC (n=24) and 8.4 with a range of 0 to 32.4 for children with NDTC (n=70). Median values for the left colon were 37.2 with a range of 0 to 110.4in children with PSTC (n=24) and 7.2 with a range of 0 to 36.0 in children with NDTC (n=70) whereas median values for the rectosigmoid were 116.4 (range 49.2 to 226.8) for PSTC children (n=24) and 27.0 (range 0 to 90.0) for NDTC children (n=70).
Daytime soiling was present in significantly more children with PSTC (n=24) compared to children with NDTC (n=70), (92% versus 69%; P = 0.05). Night time soiling was also present in significantly more children with PSTC compared to children with NDTC (17 [71%] versus 8 [11%]; P < 0.01). Daytime soiling episodes per week were significantly more frequent in children with PSTC (n=24) compared to children with NDTC (n=70), (median 14.0, range 0 to 7 versus median 5.0 range 0 to 56; P < 0.01). Night-time soiling episodes per week were also significantly more frequent in children with PSTC (n=24) compared to children with NDTC (n=70) (median 7, range 0 to 7 versus median 0, range 0 to 7; P < 0.01).
Stools were normal in significantly more children with PSTC compared to children with NDTC (75% versus 49%; P = 0.03). Pain during defecation was present in significantly more children with NDTC compared to children with PSTC (60% versus 33%; P = 0.01). Significantly more children with PSTC complained of no rectal sensation compared to children with NDTC (33% versus 14%; P = 0.03). A palpable abdominal mass was present in significantly more children with PSTC compared to children with NDTC (71% versus 39%; P = 0.02). A palpable rectal mass was present in significantly more children with PSTC compared to children with NDTC (71% versus 13%; P < 0.01). There were no significant differences between the two groups regarding: sex, age, toilet training status, age at which toilet training started, bowel movements per week, large amounts of stools every 7 to 30 days, encopresis episodes per week, abdominal pain, poor appetite or daytime or night-time urinary incontinence. The proportion of children with PSTC and rectal palpable mass, night time soiling or both was 0.34, 0.39 and 0.82 respectively. Only 7% of children without any of these characteristics had PSTC. Further analysis of the NDTC group after separation into a group with total CTT less than 63 hours and one with total CTT between 63 and 100 hours showed the same significant differences when compared with PSTC children as did the total NDTC group, allowing the merge of these children.
A case control study conducted in the Netherlands45 (1995) [EL=III] investigated the presence or absence of faecal retention in each child using CTT and compared these findings to the Barr score. The study included 211 children with complaints of infrequent defecation (paediatric constipation [PC], n=129, 64% boys, median age 8 years, range 5 to 14 years), encopresis and/or soiling (ES) (n=54, 81% boys, median age 9 years, range 5 to 17 years) or recurrent abdominal pain (RAP) (n=23, 39% boys, median age 9 years, range 5 to 16 years). Of these, 206 children underwent CTT with radiopaque markers assessed with the Metcalf method and these were compared to a plain abdominal radiograph read using the Barr score. Data on assessment of plain abdominal radiographs using Barr score was available for 101 children only. Five patients of the 211 originally recruited were excluded from the study: 4 were not able to swallow the capsules and 1 had an ‘uninterpretable’ abdominal radiography.
The total CTT (in hours) was significantly longer for children with encopresis only compared to children with RAP (mean 41.4, range 16.6 to 104.4 versus mean 32.5, range 4.8 to 69.6; P = 0.03). There were no significant differences for the CTT between children with PC (mean 79.3, range 2.4 to 384) and the other two groups. Transit time in the right colon (in hours) was significantly longer in children with PC compared to children with encopresis only (mean 13.2, range less than 1.2 to 60 versus mean 7.9, range less than 1.2 to 26.4; P < 0.01) and to children with RAP (mean 13.2 range less than 1.2 to 60, versus mean 7.7, range 1.2 to 21.6; P < 0.01). There were no significant differences between children with encopresis only and children with RAP.
Transit time in the left colon (in hours) was significantly longer in children with PC compared to children with encopresis only (mean 16.1, range less than 1.2 to 110.4 versus mean 6.8, range less than1.2 to 25.2; P < 0.01) and to children with RAP (mean 16.1, range less than1.2 to 110.4 versus mean 7.0, range 1.2 to 25.2; P < 0.01). There were no significant differences between children with encopresis only and children with RAP. Transit time in the rectosigmoid (in hours) was significantly longer in children with PC compared to children with encopresis only (mean 49.7, range less than 1.2 to 226.8 versus mean 26.7, range 4.8 to 93.6; P < 0.01) and to children with RAP (mean 49.7, range 1.2 to 226.8 versus mean 8.9, range 1.2 to 49.2; P < 0.01). It was also significantly longer in children with encopresis only compared to children with RAP (mean 26.7, range 4.8 to 93.6; P < 0.01 versus mean 8.9, range 1.2 to 49.2; P < 0.01; P = 0.05).
The interobserver agreement for the CTT was perfect in 62% of the readings of the first radiograph and a difference of one marker was present in 25%. For the second radiograph a perfect agreement was achieved in 92% of the readings and a difference of one marker was present in 6%. Sixty percent of children with PC (n=57) had mean Barr scores of 10 or more (mean of two observers) in the first radiograph and 63% in the second one. Forty-seven percent of children with isolated ES (n=30) had mean Barr scores of 10 or more in the first radiograph and 60% in the second one. Forty-seven percent of children with RAP (n=14) had mean Barr scores of 10 or more (mean of two observers) in the first radiograph and 63% in the second one. The interobserver agreement for the Barr score (the agreement between the two observers for the different segments on the same radiograph) varied from fair (k = 0.28) to moderate (k = 0.60). The intraobserver agreement (regarding the difference in quantity and quality of stool between radiograph I and II as scored by the same radiologist) varied from poor (k = 0.05) to moderate (k = 0.47) for both observers. The intraobserver agreement regarding the existence of constipation as measured by a Barr score of 10 or more points between radiographs I and II was fair for both observers (k = 0.22 and 0.25 respectively). The correlation between a positive Barr score (10 or more) and a delayed total CTT (more than 62 hours) was fair (k = 0.22) for all children. K values on a separated analysis for each group were: 0.20 (PC group), 0.02 (ES group) and 0.46 (RAP group). Abnormal Barr scores were found in at least 46% of patients with normal transit times, whereas positive Barr scores correlated only with a total CTT exceeding 100 hours.
A diagnostic prospective case series conducted in the UK46 (1994) [EL=III] assessed the reliability of interpretation and the clinical value of solid marker transit studies in children with soiling and spurious diarrhoea, otherwise known as overflow incontinence. Fifty-two children with a median age of 8 years (range 2 to 13.5 years) with constipation and/or soiling underwent CTT with radiopaque markers. No reference tests were used but outcomes of CTT were related to the frequency of bowel movements and soiling. In relation to the patterns of transit time 21 children (40%) were diagnosed with normal transit, 4 children (8%) with mild delay, 9 children (17%) with moderate delay and 18 children (35%) with severe delay. In relation to the patterns of marker distribution 15 children (29%) were diagnosed with pancolonic transit delay, 5 children (10%) with segmental transit delay and 11 children (21%) with outlet obstruction.
Significantly more children with severe transit delay (n=18) had fewer than two bowel movements per week when compared to children with normal transit (n=21), (87% versus 27%; P < 0.001). Significantly more children with severe transit delay (n=18) had more than three soiling episodes per week when compared to children with normal transit (n=21); (92% versus 35%; P < 0.005). No correlation was found between the duration of the symptoms and the severity of transit delay. Thirty-nine percent of the children with severe delay (n=18) had outlet obstruction, 56% pancolonic transit delay and 5% segmental transit delay (in descending colon). Significantly more children with mild delay (n=4) had segmental transit delay (in rectosigmoid) than pancolonic transit delay (75% versus 25%; P < 0.005).
Significantly more children with outlet obstruction had fewer than two bowel movements per week compared to children with segmental transit delay (100% versus 83%; P < 0.05). Significantly more children with pancolonic transit delay had fewer than two bowel movements per week compared to children with segmental transit delay (83% versus 33%; P < 0.05). There were no significant differences between children with outlet obstruction and children with pancolonic transit delay. Significantly more children with outlet obstruction had more than three soiling episodes per week compared to children with segmental transit delay (100% versus 0%; P < 0.05). Significantly more children with pancolonic transit delay had more than 3 soiling episodes per week compared to children with segmental transit delay (57% versus 0%; P < 0.05). The interobserver coefficient of variation was 2.1% and the intraobserver coefficient of variation was 3.1%.
A diagnostic case control study conducted in Italy47 (1994) [EL=III] studied colonic transit and anorectal motility in children with severe brain damage, looking for differences from asymptomatic children and from patients with functional faecal retention and normal neurologic development. The study included 42 children. Patients were 16 children with brain damage referred for gastroenterologic evaluation of constipation (10 boys, mean age 5.1 ± 3.5 years, range 1.5 to 12 years). Controls were 15 children diagnosed with idiopathic constipation (IC, termed functional faecal retention in the paper) (9 boys, mean age 6.0 ± 2.9 years, range 2 to 11 years) and 11 children with no gastrointestinal problems (7 boys, mean age 5.6 ± 3.9 years, range 2 to 12 years). All children underwent total gastrointestinal transit time (TGITT)* with radiopaque markers.
The TGITT (in hours) was not significantly different in children with brain damage compared to children with functional faecal retention (mean 106.4 ± 6.1 versus 98.6 ± 5.1). The total number of markers at 48 hours and 72 hours (mean,) in the left colon was significantly larger in brain damaged children compared to children with IC (at 48 hrs: mean 7.3 ± 1.3 standard error of the mean (SEM) versus mean 3.0 ± 1.0 SEM; P < 0.05), (at 72 hrs: mean 3.3 ± 0.8 SEM versus mean 0.5 ± 0.3 SEM; P < 0.01). The distribution of the markers in both right colon and rectum was not significantly different between the two groups at any time. Twenty-nine of the children originally undergoing evaluation for severe brain damage were found to have constipation, but only 16 were included in the study. It is not clear why the other 13 were excluded. Exact values for all segmental transit times in the two groups were not reported.
A multicentre retrospective case series conducted in Switzerland48 (1993) [EL=III] investigated the relationship between clinical, manometric and histological findings in a group of children with chronic constipation in order to evaluate the role of anorectal manometry in the diagnosis of neuronal intestinal dysplasia and the relationship of histological and manometric findings to clinical severity of constipation and outcome. Forty-eight children (25 boys, mean age 6.4 years ± 5.2) with initial symptoms of chronic constipation or soiling, or obstructive symptoms in early life suggestive of Hirschsprung's disease, were included in the study. Thirty children underwent CTT with radiopaque markers. The mean total transit time for children with normal histology (n=15) was 70.0 hours ± 42.6.The results for segmental transit times were not reported and it is not clear whether they were measured. CTT results for children diagnosed with abortive and classic neuronal intestinal dysplasia are not reported for the purposes of this review as they are considered organic causes of constipation.
A diagnostic retrospective case series conducted in France49 (1998) [EL=III] analysed epidemiologic, manometric and radiologic data in a large population of young patients presenting in a paediatric tertiary care hospital in order to classify different types of idiopathic constipation according to age of onset, sex and pelvic floor function. The study included 1182 children (63% boys) diagnosed with constipation with or without encopresis. Children were divided into two groups: constipated children without encopresis (n=855) and constipated children with encopresis (n=327). Sixty-five percent of the patients without encopresis were younger than 4 years. Of the children, 378 underwent CTT with radiopaque markers. No other test was used as a comparator.
The total CTT (in hours) was significantly longer in patients with encopresis (n=168) and patients without encopresis age over 4 years (n=112) and under 4 years (n=77) compared to controls (n=21) (median 67.2, range 2 to 168 versus median 54.6, range 9 to 168 versus median 49.6, range 8 to 161 versus median 22.8, range 9.4 to 56.4; P < 0.0001). Patients with encopresis had significantly longer total CTT compared to patients without encopresis age over 4 years (median 67.2, range 2 to 168 versus median 54.6, age 9 to 168; P < 0.05).
Transit time in the right colon (in hours) was significantly longer in patients without encopresis age over 4 years and under 4 years compared to controls (median 12, range 0 to 48 and median 14.8, range 0 to 96 versus median 7.2, range 0.6 to 19.2; P < 0.0005) and also in patients with encopresis compared to controls (median 14, range 0 to 144 versus median 7.2, range 0.6 to 19.2; P < 0.0001). Transit time in the left colon (in hours) was significantly longer in patient without encopresis age over 4 years and under 4 years and in patients with encopresis compared to controls (median 12, range 0 to 96 and median 12.4, range 0 to 72 and median 13.6, range 0 to 96 versus 7.4 (1.2 to 22.8); P < 0.005). Transit time in the rectosigmoid (in hours) was significantly longer in patients without encopresis age over 4 years and patients with encopresis compared to controls (median 26.4, range 0 to 108 and median 30.2, range 0 to 142 versus median 10.4, range 1.21 to 34.2; P < 0.0001) and also when comparing patients without encopresis age under 4 years with controls (median 18.4, range 0 to 106 versus median 10.4, range 1.21 to 34.2; P < 0.005). Transit time (in hours) in the total colon plus the rectum was significantly longer in all patient groups compared to controls (median 49.6, range 8 to 161, median 54.6, range 9 to 168 and median 67.2, range 2 to 168 versus 22.8 (9.4 to 56.4); P < 0.0001). Transit time in the total colon plus the rectum was significantly longer in patients with encopresis patients compared to patients without encopresis age over 4 years (median 67.2, range 2 to 168 versus median 54.6, range 9 to 168; P < 0.05).
Of the total sample, 29% was diagnosed with normal transit. Significantly more patients with encopresis were diagnosed with normal transit compared to patients without encopresis age under 4 years (n=38 (10.6%) versus n=33 (9.2%); P < 0.001). Of the total sample, 36% was diagnosed with terminal constipation, which is defined as delay in the rectosigmoid site with or without delay in the right or left colon. Significantly more patients without encopresis age over 4 years were diagnosed with terminal constipation compared to those under 4 years (n=42 (37.5%) versus n=17 (22%); P < 0.05). Significantly more patients with encopresis were diagnosed with terminal constipation compared to patients without encopresis age under 4 years (n=70 (41.5%) versus n=17 (22%); P < 0.005). Twenty-three percent of the total sample was diagnosed with non-terminal constipation and 12% with pancolic constipation.
A diagnostic case–control study conducted in Italy50 (1985) [EL=III] quantified bowel function in healthy children in terms of frequency of defecation, gastrointestinal transit time and manometric characteristics of the anorectal tract and compared variables of bowel function in children with chronic constipation with those in the normal population. The study included 166 children of whom 63 were patients with long-standing constipation (mean age 5.4 years ± 4.1, range 2 months to 4 years), and 103 were controls who were healthy children free of bowel complaints. Total gastrointestinal transit time (TGITT) was measured with radiopaque markers in all children and this was related to the frequency of defecation.
The mean TGITT (in hours) for the healthy controls was 25.0 ± 3.7 with a range of 19 to 33. Fifty-three patients had a TGITT of more than 33 hours and 10 patients had a TGITT more than 33 hours. Segmental transit time was measured in 39 out of 53 children with prolonged transit time and it was lowest in the colon for three patients, in the rectum for 24 patients and in the colon and rectum for 12 patients. The stool frequency and the TGITT were significantly correlated in patients with prolonged transit time and in healthy controls (patients with TGITT more than 33 hours (n=53 had a mean of 2.5 ± 0.9; r=0.75, P < 0.001 and healthy controls (n=78) had a mean of 6.3 ± 1.3; r=0.78, P < 0.001). In 7 of 53 patients with TGITT more than 33 hours, the bowel frequency overlapped the range observed in the control subjects. Segmental colonic transit times (right and left colon and rectosigmoid) were evaluated but results were not reported.
A diagnostic case–control study conducted in Italy51 (1984) [EL=III] determined the motility characteristics of the anorectum and measured TGITT in children with chronic constipation, with or without faecal overflow. The study included 99 children, of which 53 were patients with constipation of several months of duration with or without soiling (40 boys, mean age 8.3 years, range 4.8 to 12.9). Controls were 46 healthy children without gastrointestinal complaints (24 boys, mean age 8.1 years, range 4.2 to 12). Controls were matched for age and weight but not for sex with the constipated children. All children underwent TGITT with radiopaque markers. No test was used as a comparator.
The TGITT (in hours) was significantly longer in patients with soiling (n=32) compared to the healthy controls (mean 58 ± 14.3, range 36 to 86 versus mean 25.6 ± 3.7, range 19 to 33; P < 0.001). It was also significantly longer in patients without soiling (n=21) compared to the healthy controls (mean 61.1 ± 15, range 36 to 96 versus mean 25.6 ± 3.7, range 19 to 33; P < 0.001). Segmental transit times were not measured.
A diagnostic prospective case series conducted in France52 (1983) [EL=III] described the clinical presentation of children with idiopathic disorders of faecal continence and aimed to demonstrate that they have functional abnormalities of large bowel motility. The study included 176 patients aged 2 to 15 years (64% boys) with idiopathic disorders of bowel function other than Hirschsprung's disease. All patients underwent CTT with radiopaque markers. The transit time of one radiopaque marker in all three colonic segments was significantly longer in constipated children (with or without spina bifida occulta) compared to normal children (ascending colon: mean 13 hours 24 minutes ± 1 hour 5 minutes versus mean 7 hours 10 minutes ± 1 hour 4 minutes; P < 0.05), (descending colon: mean 13 hours 49 minutes ± 1 hour 37 minutes versus mean 7 hours 37 minutes ± 1 hour 3 minutes; P < 0.05) and (rectum: 30 hours 22 minutes ± 2 hours 42 minutes versus 11 hours 4 minutes ± 1 hour 5 minutes; P < 0.05). There were no significant differences regarding segmental transit times between children with and without spina bifida occulta. Total transit times were not reported.