Abstract
Background
The aims of this study were to determine prevalence, risk factors and treatment of constipation in patients with Cystic Fibrosis (CF), as well as the diagnostic value of abdominal radiography.
Methods
A cohort of 214 pediatric CF patients was investigated. Furthermore, 106 abdominal radiographs of CF patients with or without constipation were independently assessed by three observers on two separate occasions using the Barr and Leech scores.
Results
The prevalence of constipation was 47%. Low total fat absorption and meconium ileus were independent risk factors for constipation in CF, while fiber and fluid intake were not associated. In CF patients the inter and intraobserver variabilities of the Barr and Leech scores were poor to moderate.
Conclusion
Constipation is a significant medical issue in CF and was associated with low total fat absorption and a history of meconium ileus. Finally, abdominal radiography seems of little value in the regular follow-up of CF patients.
Keywords
1. Introduction
Constipation is one of the gastrointestinal manifestations of Cystic Fibrosis (CF). It is characterized by a reduced stool frequency and increased consistency, usually in combination with abdominal pain and distension and generally responds well to conservative medical treatment [
1
, Houwen RH, van der Doef HP, Sermet I, Munck A, Hauser B, Walkowiak J, et al. The ESPGHAN Cystic Fibrosis Working Group: Defining DIOS and Constipation in Cystic Fibrosis with a multicenter study on the incidence, characteristics and treatment of DIOS. J Pediatr Gastroenterol Nutr. In press doi:10.1097/MPG.0b013e3181a6da59.
2
]. The frequency of this condition in CF is unclear. The only study addressing this problem so far reported a prevalence of 26% in patients aged 0–20 years [[2]
].The main etiological factor for constipation in CF patients seems to be an altered intestinal fluid composition, caused by a defective expression of the Cystic Fibrosis transmembrane regulator (CFTR) protein in the gut [
3
, 4
]. In addition it is generally thought that a more aggressive treatment with pancreas supplements would result in more compact feces leading to fecal impaction [5
, 6
], although no correlation between pancreas supplement dose and constipation was found [[7]
].Key elements in the diagnosis of constipation are a careful medical history and physical examination. In addition abdominal radiography is frequently performed when constipation is suspected. In this respect several scoring systems, like the Barr [
[8]
] and the Leech [[9]
] scores, are available to assess the severity of fecal impaction. However the diagnostic value of these scores has only been investigated in patients with functional constipation, but not in CF.Recently the ESPGHAN CF Working Group made a strict distinction between the (sub)acute complete ileocecal obstruction, as seen in the distal intestinal obstruction syndrome (DIOS) and the gradual fecal impaction of the total colon in constipation [
[1]
]. DIOS patients are treated generally successfully with intensive laxative treatment (meglumine diatrizoate enema, polyethylene glycol lavage, oral laxatives or an enema) [Houwen RH, van der Doef HP, Sermet I, Munck A, Hauser B, Walkowiak J, et al. The ESPGHAN Cystic Fibrosis Working Group: Defining DIOS and Constipation in Cystic Fibrosis with a multicenter study on the incidence, characteristics and treatment of DIOS. J Pediatr Gastroenterol Nutr. In press doi:10.1097/MPG.0b013e3181a6da59.
[1]
]. It seems logical that in constipated CF patients generally a milder laxative regime will be used, but no such data are available at present.Houwen RH, van der Doef HP, Sermet I, Munck A, Hauser B, Walkowiak J, et al. The ESPGHAN Cystic Fibrosis Working Group: Defining DIOS and Constipation in Cystic Fibrosis with a multicenter study on the incidence, characteristics and treatment of DIOS. J Pediatr Gastroenterol Nutr. In press doi:10.1097/MPG.0b013e3181a6da59.
The aims of this study were therefore to determine the prevalence of constipation, its risk factors and diagnostic value of abdominal radiography in patients with CF. In addition we listed laxative treatment used.
2. Methods
2.1 Prevalence and risk factors
A retrospective cohort study of all pediatric CF patients (age ≤18 years) under treatment at the University Medical Center Utrecht, The Netherlands on January 1st, 2006 was performed according to the guidelines of the medical ethics board of the University Medical Center Utrecht, The Netherlands.
For constipation and DIOS the recently published definitions of the ESPGHAN CF Working Group were used [
[1]
]. Constipation was defined as [1] abdominal pain and/or distension or [2a] a reduced frequency of bowel movements in the last few weeks and/or [2b] increased consistency of stools in the last few weeks, while [3] the symptoms are relieved by the use of laxatives. DIOS was defined as the combination of [1] complete intestinal obstruction, as evidenced by vomiting of bilious material and/or fluid levels in small intestine on an abdominal radiograph with [2] a fecal mass in ileocecum and [3] abdominal pain and/or distension. Incomplete or impending DIOS was defined as (1) a short history (days) of abdominal pain or distension or both and (2) a fecal mass in ileocecum, but without signs of complete obstruction.Houwen RH, van der Doef HP, Sermet I, Munck A, Hauser B, Walkowiak J, et al. The ESPGHAN Cystic Fibrosis Working Group: Defining DIOS and Constipation in Cystic Fibrosis with a multicenter study on the incidence, characteristics and treatment of DIOS. J Pediatr Gastroenterol Nutr. In press doi:10.1097/MPG.0b013e3181a6da59.
Three-day dietary records (3 consecutive days including 1 weekend day) were completed by patient and/or family every year and analyzed by a registered dietitian (J.W.). The dietary data used for analysis were obtained 0 to 6 months before patients presented for the first time with constipation and were compared with the last available dietary data obtained in the patients without constipation or DIOS (complete or incomplete). Nutrient intake was expressed as percentage of the gender- and age-specific Reference Daily Intake (RDI) [
[10]
]. Total fat absorption was calculated from the mean daily fat intake of three-day dietary records and the daily fecal fat output and was expressed as percentage of the mean daily fat intake.For the determination of CFTR genotypes only alleles with known mutations were analyzed. Then the CFTR genotypes of the CF patients were subdivided into two groups: the first group consisted of patients with a severe genotype, defined as 2 severe CFTR mutations (class I–III) and the second group consisted of patients with a mild genotype, defined as at least 1 mild mutation (class IV–V) [
[11]
]. The distribution of DF508 homozygous patients was also examined.2.2 Abdominal radiography
Between April and December 2006 all pediatric CF patients (age ≤18 years) who visited the outpatient clinic for the annual checkup in the University Medical Center Utrecht underwent abdominal radiography regardless of the presence or absence of abdominal symptoms or constipation. A retrospective analysis of this group was subsequently performed according to the guidelines of the medical ethics board of the University Medical Center Utrecht, The Netherlands. Three observers; a medical student (S.F.), an experienced pediatric radiologist (F.B.) and an experienced pediatric gastroenterologist (F.K.) independently assessed the abdominal radiographs taken. The three observers were blinded to the study objective and each abdominal radiograph was evaluated on two separate occasions, 3 weeks apart. Each abdominal radiograph was scored according to two different scoring systems and before the first scoring, the different systems were not discussed by the observers. The first method, described by Barr et al. [
[8]
], quantifies the amount of feces in four different bowel segments (ascending colon, transverse colon, descending colon and rectum) and is scored respectively from 0 to 2, 0 to 5, 0 to 5 and 0 to 5. Also the consistency of the feces, i.e. granular or rocky stools is scored respectively from 0 to 3 and 0 to 5. Constipation is defined as a score ≥10. The second method was described more recently by Leech et al. [[9]
]. In this system the colon is divided into three colonic segments (the right colon, the left colon and the rectosigmoid segment) and the amount of feces in each segment is scored from 0 to 5. Constipation is defined as a score ≥9.The presence of radiological constipation in the first evaluation of all three observers according to the Barr and Leech scores were compared with the presence of constipation according to the ESPGHAN criteria [
[1]
] as gold standard and the sensitivity, specificity, positive and negative predictive values (PPV and NPV) were calculated. Furthermore, the interobserver variability of the Barr and Leech scores was calculated using the first evaluation of the three observers and the intraobserver variability was calculated using both evaluations of the observers.Houwen RH, van der Doef HP, Sermet I, Munck A, Hauser B, Walkowiak J, et al. The ESPGHAN Cystic Fibrosis Working Group: Defining DIOS and Constipation in Cystic Fibrosis with a multicenter study on the incidence, characteristics and treatment of DIOS. J Pediatr Gastroenterol Nutr. In press doi:10.1097/MPG.0b013e3181a6da59.
2.3 Statistical analysis
Data were described as mean and standard deviation for ordinal values, and absolute and relative frequencies for nominal values. Logistic regression was used to test the effect of potential risk factors and variables contributing significantly (p<0.05) were included in the multivariate analysis. Unweighted kappa coefficients were calculated as indicators of inter and intraobserver variabilities for nominal variables (presence or absence of radiological constipation) and weighted kappa coefficients were calculated for ordinal variables (amount of points scored by the observers). Kappa coefficients <0.20, 0.21–0.40, 0.41–0.60, 0.61–0.80, and 0.81–1.00 were considered to indicate poor, fair, moderate, good and very good agreement, respectively [
[12]
]. Values were considered significant if p≤0.05. The weighted kappa coefficient was calculated using R software (Free Software Foundation Inc., Boston, MA, USA), while all other statistical analyses were performed using SPSS software (SPSS Inc., Chicago, IL, USA).3. Results
3.1 Prevalence and risk factors
The study group consisted of 230 pediatric CF patients (age ≤18 years) under treatment at the University Medical Center Utrecht, The Netherlands on January 1st, 2006. Within this study group, 107 patients (47%) had a history of constipation, while 46 patients (20%) were constipated at January 1st, 2006.
Sixteen patients with a history of DIOS (complete or incomplete) were excluded in the analysis of risk factors. The characteristics of the 107 constipation patients were compared with 107 CF patients without a history of constipation or DIOS (complete or incomplete) and are reported in Table 1.
Table 1Clinical characteristics associated with constipation in Cystic Fibrosis.
| Characteristics | All patients | Constipation | Controls | p-value | OR (95%CI) |
|---|---|---|---|---|---|
| Patient number | 214 | 107 | 107 | ||
| Gender (male) | 119 (56%) | 61 (57%) | 58 (54%) | 0.68 | 1.10 (0.65–1.92) |
| Age diagnosis CF | 1.31 (2.08) | 1.12 (1.64) | 1.49 (2.43) | 0.20 | 0.92 (0.80–1.05) |
| Current age | 9.96 (4.64) | 10.31 (4.37) | 9.60 (4.90) | 0.26 | 1.03 (0.98–1.10) |
| CFTR genotype | |||||
| Severe | 174 (81%) | 93 (87%) | 81 (76%) | 0.34 | 1.64 (0.60–4.51) |
| DF508/DF508 | 130 (61%) | 73 (68%) | 57 (53%) | 0.25 | 1.83 (0.66–5.11) |
| Mild | 17 (8%) | 7 (7%) | 10 (9%) | ||
| Clinical manifestations | |||||
| Pancreas insufficiency | 207 (97%) | 106 (99%) | 101 (94%) | 0.091 | 0.16 (0.019–1.34) |
| Meconium ileus | 19 (9%) | 14 (13%) | 5 (5%) | 0.038 | 3.07 (1.07–8.86) |
| Dietary intake | |||||
| Fiber (% RDI) | 0.58 (0.19) | 0.58 (0.23) | 0.58 (0.17) | 0.93 | 0.92 (0.14–6.27) |
| Fluid (% RDI) | 0.87 (0.21) | 0.89 (0.21) | 0.86 (0.21) | 0.46 | 1.88 (0.35–10.00) |
| Total fat absorption | 0.89 (0.08) | 0.86 (0.09) | 0.90 (0.07) | 0.012 | 0.003 (0.00–0.28) |
a Mean (SD).
b Severe genotype or DF508 homozygous vs. mild genotype.
Meconium ileus was significantly more frequent in patients with a history of constipation than in patients without constipation (13% vs. 5%, p=0.038). In the subgroup of the 19 meconium ileus patients surgical treatment for meconium ileus was more common in constipation patients than in patients without constipation (71% vs. 60%), although no statistical significance was reached (p=1.00). Also pancreatic insufficiency was more common in the constipated group than in control CF patients (99% vs. 94%), although this difference was not significant (p=0.091). However total fat absorption was significantly lower in patients with constipation than in patients without constipation (0.86±0.09 vs. 0.90±0.07, p=0.012). All other variables (current age, age at diagnosis of CF, gender, CFTR genotype, mean fiber intake and mean fluid intake) were not significantly different between patients with or without constipation.
Logistic regression analysis showed that meconium ileus at birth (p=0.024; OR 4.69, 95%CI 1.22–18.0) and low total fat absorption (p=0.010; OR 0.002, 95%CI 0.000–0.24) were indeed both independently associated with constipation.
3.2 Abdominal radiography
Abdominal radiography was performed in the 106 CF patients who visited the outpatient clinic for the annual checkup in the University Medical Center Utrecht, The Netherlands between April and December 2006. Of the 106 patients 36 (34%) were constipated and 70 (66%) were not constipated according to the ESPGHAN criteria [
[1]
]. With this criterion as a gold standard sensitivity, specificity, PPV and NPV of radiological constipation according to the different scoring systems generally was low (Table 2). For the Barr score the observers reported sensitivities ranging from 0.14 to 0.61, specificities ranging from 0.43 to 0.96, PPV ranging from 0.35 to 0.63 and NPV ranging from 0.68 to 0.71. For the Leech score the observers report sensitivities ranging from 0.11 to 0.72, specificities ranging from 0.34 to 0.93, PPV ranging from 0.36 to 0.44 and NPV ranging from 0.67 to 0.71.Houwen RH, van der Doef HP, Sermet I, Munck A, Hauser B, Walkowiak J, et al. The ESPGHAN Cystic Fibrosis Working Group: Defining DIOS and Constipation in Cystic Fibrosis with a multicenter study on the incidence, characteristics and treatment of DIOS. J Pediatr Gastroenterol Nutr. In press doi:10.1097/MPG.0b013e3181a6da59.
Table 2Sensitivity, specificity and positive and negative predictive values of the presence of radiological constipation according to the Barr and Leech scores in children with Cystic Fibrosis.
| Radiologist | Gastroenterologist | Medical student | |
|---|---|---|---|
| Barr score | |||
| Sensitivity | 0.61 | 0.14 | 0.53 |
| Specificity | 0.43 | 0.96 | 0.59 |
| Positive predictive value | 0.35 | 0.63 | 0.40 |
| Negative predictive value | 0.68 | 0.68 | 0.71 |
| Leech score | |||
| Sensitivity | 0.72 | 0.11 | 0.50 |
| Specificity | 0.34 | 0.93 | 0.63 |
| Positive predictive value | 0.36 | 0.44 | 0.41 |
| Negative predictive value | 0.71 | 0.67 | 0.71 |
Furthermore, the inter and intraobserver variabilities of the three observers according to the two different scoring systems generally were low too. The inter and intraobserver variabilities of the Barr and Leech scores for the presence of radiological constipation (Barr score <10 vs. ≥10 and Leech score <9 vs. ≥9) ranged from an unweighted kappa coefficient of 0.08 (poor) to 0.44 (moderate) and the inter and intraobserver variabilities of the Barr and Leech scores for the amount of points scored ranged from a weighted kappa coefficient of 0.09 (poor) to 0.55 (moderate).
3.3 Treatment of constipation patients
In our patient group 58% had had at least 1 oral laxative (OL), 26% 2 OL, 8% 3 OL, 6% 4 OL and 2% 5 OL. Generally patients started with lactulose or polyethylene glycol. If the effect was insufficient one or two additional OLs were added. In 53 patients (50%) at least once an enema was necessary and in 14 patients (13%) intestinal lavage. Eight patients received a stimulant laxative for a short period.
4. Discussion
In this study we determined the prevalence, risk factors and treatment of constipation in a cohort of CF patients, as well as the diagnostic value of abdominal radiography in this condition.
One hundred and seven out of 230 patients (47%) had a history of constipation, while 46 out of 230 patients (20%) were constipated at January 1st, 2006. Prevalence numbers of constipation in CF are scarce; only one study published in 1986 has reported prevalence numbers of constipation in CF [
[2]
]. This study observed that constipation had been present in 26% of all CF patients aged 0–20 years [[2]
], which is significantly lower than prevalence in the present study (p<0.001). While it is possible that the prevalence of constipation has increased over time, it is as likely that laxatives are prescribed more easily nowadays in CF patients suspected of constipation, especially as current laxatives are almost devoid of side effects. As both in our definition for constipation and in the definition of Rubinstein et al. [[2]
] the use of laxatives is a key component, such a change in practice might result in the increasing prevalence numbers we here describe.In the current study we found that meconium ileus was independently associated with constipation. An association between meconium ileus and DIOS has been reported previously [
[13]
]. It seems indeed logical that both meconium ileus, DIOS and constipation in CF are an interrelated group of manifestations, ranging from severe to mild intestinal obstruction and sharing a common pathophysiology.The relationship between pancreatic insufficiency or poorly controlled steatorrhea and constipation is unclear and conflicting results have been published. In general, it is thought that constipation is correlated with highly dosed pancreatic supplements [
5
, 6
]. However this is not supported by Baker et al. [[7]
], who report no correlation between constipation and the dosage of pancreatic supplements. We now found that constipation patients have a lower total fat absorption than control patients, although both patient groups (with and without constipation) had an adequate control of steatorrhea with a mean total fat absorption of 86% and 90% respectively [[14]
]. Slow intestinal transit and malabsorption may allow undigested food to enter the colon over a prolonged period [15
, 16
]. This could lead, in combination with impaired intestinal secretion [3
, 4
], to sticky intestinal mucus and constipation.Finally, in concordance with a report describing Belgian CF patients [
[17]
], fiber intake was not correlated with constipation in CF. Furthermore, no differences in the fluid intake between patients with or without constipation were found, despite the general opinion that inadequate fluid intake is an etiological factor of constipation in CF [[2]
].In children with CF the Barr and Leech scores have poor sensitivity, specificity, PPV and NPV for diagnosing constipation, with a considerable overlap in Barr and Leech scores between constipated and nonconstipated patients. In addition, poor inter and intraobserver variabilities were found when scoring abdominal radiographs in CF patients. Similar results have been published in patients with functional constipation; a systematic review showed a low diagnostic value of abdominal radiography (sensitivity 60–80% and specificity 35–90%) [
[18]
], while the inter and intraobserver variabilities of the different scoring systems are poor too [19
, 20
]. Consequently, abdominal radiography is not recommended as a standard diagnostic tool in the regular gastrointestinal follow-up of CF patients. However, abdominal radiography is useful to differentiate between constipation and the distal intestinal obstruction syndrome in CF patients with acute abdominal pain [[1]
].Houwen RH, van der Doef HP, Sermet I, Munck A, Hauser B, Walkowiak J, et al. The ESPGHAN Cystic Fibrosis Working Group: Defining DIOS and Constipation in Cystic Fibrosis with a multicenter study on the incidence, characteristics and treatment of DIOS. J Pediatr Gastroenterol Nutr. In press doi:10.1097/MPG.0b013e3181a6da59.
Currently, polyethylene glycol seems to be the preferred initial treatment for both constipation and DIOS (complete and incomplete), because it is as effective and does not have the side effects that are inherent to lactulose (flatulence and abdominal cramps) [
[21]
].In conclusion, constipation is a significant medical issue in CF patients, with a prevalence of 47%. Furthermore, we found that low total fat absorption and meconium ileus were independent risk factors for constipation in CF, while fiber and fluid intake were not associated with constipation in CF. In addition, the diagnostic value of abdominal radiography in CF is limited. Abdominal radiography is therefore not recommended in the regular follow-up of CF patients.
Acknowledgments
The authors wish to thank S. van Zuilen, secretary of the Wilhelmina Children's Hospital, University Medical Center Utrecht, Utrecht, The Netherlands for her contribution to the database construction and wish to thank J. van der Laag of the Wilhelmina Children's Hospital, University Medical Center Utrecht, Utrecht, The Netherlands for his contribution to the discussion.
The study was funded by the Wilhelmina Research Fund, grant number OZF 2005/04. This organization had no involvement in the study design, in the collection and interpretation of data, in the writing of the manuscript and in the decision to submit the manuscript for publication.
References
Houwen RH, van der Doef HP, Sermet I, Munck A, Hauser B, Walkowiak J, et al. The ESPGHAN Cystic Fibrosis Working Group: Defining DIOS and Constipation in Cystic Fibrosis with a multicenter study on the incidence, characteristics and treatment of DIOS. J Pediatr Gastroenterol Nutr. In press doi:10.1097/MPG.0b013e3181a6da59.
- Constipation and meconium ileus equivalent in patients with cystic fibrosis.Pediatrics. 1986; 78: 473-479
- Relationship between intestinal function and chloride secretion in patients with cystic fibrosis.Neth J Med. 1992; 41: 110-114
- The DeltaF508 mutation results in loss of CFTR function and mature protein in native human colon.Gastroenterology. 2004; 126: 32-41
- Diagnosis and treatment of intestinal malabsorption in cystic fibrosis.Pediatr Pulmonol. 2006; 41: 35-49
- Nutrition in patients with cystic fibrosis: a European Consensus.J Cyst Fibros. 2002; 1: 51-75
- Pancreatic enzyme therapy and clinical outcomes in patients with cystic fibrosis.J Pediatr. 2005; 146: 189-193
- Chronic and occult stool retention: a clinical tool for its evaluation in school-aged children.Clin Pediatr (Phila). 1979; 18: 676-679
- Evaluation of a method of assessing faecal loading on plain abdominal radiographs in children.Pediatr Radiol. 1999; 29: 255-258
Health Council of The Netherlands. Dietary Reference Intakes: energy, proteins, fats and digestible carbohydrates. The Hague: Health Council of The Netherlands, 2001; publication no. 2001/19.
- Molecular mechanisms of CFTR chloride channel dysfunction in cystic fibrosis.Cell. 1993; 2: 1251-1254
- Practical statistics for medical research. Chapman and Hall, London1991: 403-409
- Relative contribution of genetic and nongenetic modifiers to intestinal obstruction in cystic fibrosis.Gastroenterology. 2006; 131: 1030-1039
- Indirect pancreatic function tests in children.J Pediatr Gastroenterol Nutr. 2005; 40: 107-114
- Prolonged small-intestinal transit time in cystic fibrosis.Br Med J. 1983; 287: 1011-1013
- Intestinal permeability to 51Cr-EDTA and orocecal transit time in cystic fibrosis.J Pediatr Gastroenterol Nutr. 1992; 14: 204-207
- Evaluation of dietary fiber intake in Belgian children with cystic fibrosis: is there a link with gastrointestinal complaints?.J Pediatr Gastroenterol Nutr. 2002; 35: 610-614
- Diagnostic value of abdominal radiography in constipated children: a systematic review.Arch Pediatr Adolesc Med. 2005; 159: 671-678
- The Leech method for diagnosing constipation: intra- and interobserver variability and accuracy.Pediatr Radiol. 2006; 36: 43-49
- Defaecation disorders in children, colonic transit time versus the Barr-score.Eur J Pediatr. 1995; 154: 277-284
- Double-blind randomized evaluation of clinical and biological tolerance of polyethylene glycol 4000 versus lactulose in constipated children.J Pediatric Gastroenterol Nutr. 2005; 41: 625-633
Article info
Publication history
Accepted:
November 5,
2009
Received in revised form:
November 4,
2009
Received:
August 18,
2009
Footnotes
☆The article was presented at the 31st European Cystic Fibrosis Conference, June 2008, Prague, Czech Republic (Abstract J Cyst Fibros 2008;7 Suppl 2:S79).
Identification
Copyright
© 2009 European Cystic Fibrosis Society. Published by Elsevier Inc.
User license
Elsevier user license | How you can reuse 
Elsevier's open access license policy
Elsevier user license
Permitted
For non-commercial purposes:
- Read, print & download
- Text & data mine
- Translate the article
Not Permitted
- Reuse portions or extracts from the article in other works
- Redistribute or republish the final article
- Sell or re-use for commercial purposes
Elsevier's open access license policy
