• 
• 

Three Part Question

In a [child presenting to the Emergency Department] how useful are [Abdominal X-rays] in the [diagnosis of constipation]?

Clinical Scenario

An eight year old boy present to the Emergency Department with acute abdominal pain. His parent tells you that the boy hasn't passed any stools for a few days. You wonder whether an abdominal x-ray would be useful in diagnosing constipation.

Search Strategy

Ovid via medline 1950 to week 3 Oct 2007 and Embase 1980 to June 2007 using OVID interface. The Cochrane Library Issue 3 2007.
[constipation.mp. or exp Constipation exp Feces/ or stool$.mp. defec$.mp. or defaec$.mp.] and [(abdom$ adj x-ray).mp abdom$ adj radiograph$.mp)]
Limit to humans and english language and("newborn infant (birth to 1 month)" or "infant (1 to 23 months)" or "preschool child (2 to 5 years)" or "child (6 to 12 years)" or "adolescent (13 to 18 years)")

Search Outcome

99 articles were found. Altogether 7 papers were suitable. One of the papers found was an analysis of the results of four of the papers found, so 3 papers were suitable.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Maurice Van Den Bosch, et al, Nethelands 2006	40 radiographs of children aged between 3 and 12 were analysed	Retrospective study reviewing 40 abdominal radiographs of children referred to a hospital for assessment due to constipation. The radiographs were assessed by two different observers on two different occasions six weeks apart. The variability between the two observers (inter-observer variability) was calculated as was the variability of an individual observers findings on the two different occasions (intra-observer variability) six weeks apart. Kappa coefficients were calculated to show variability between the results on the different occasions. A low Kappa score shows poor agreement whereas a high Kappa score shows good agreement between the observers. The highest possible score is 1.0. The radiographs were evaluated by three different methods by Barr et al in 1979, by Blethyn et al in 1995 and Leech et al in 1999	Barr score	The inter-observer variability was 0.75 and 0.66 showing good agreement between the observers. The intra-observer variability was 0.45 and 0.77. Thus the Barr score was not good at reproducing results.	The study focussed more on concordance of results between the observers, and reproducibility of results than the diagnosis of constipation in the first place. All the patients of whom the radiographs were used had symptoms of constipation
			Leech score	The Leech score had the highest reproducibility with high agreement on both the inter-observer (0.88 and 1.0) and intra-observer variability (0.91 and 0.84)
			Blethyn score	The reproducibility of the Blethyn score was the poorest of all three score, with Kappa values of 0.66 and 0.61 for inter-observer variability and 0.31 to 0.43 for intra-observer variability.
Bongers M.E.J et al, 2006 Nethelands	A systematic analysis of the role of abdominal radiography in children with functional gastrointestinal problems	Article review, reviewing all the literature that is available on use of abdominal radiographs on children with gastrointestinal problems.	Barr et al	Evaulation by Barr et al showed high inter- and intra-observer variability of 0.8 and 0.85 respectively
			Leech et al	Leech et al reported no intra-observer variability, but there was significant inter-observer variability
			De Lorijin et al assessed the Leech score	The study showed high intra-observer variability and the results for inter-observer variability was found to show large variability.
Reuchlin-Vroklage L.M, et al, 2005 Netherlands	A systematic review assessing the evidence available about the use of abdominal radiographs in constipation in children.	Systematic review	Number of articles found	Found 6 articles that met the study specifications. 2 of these articles were evaluated to be of high quality
			Conclusions	Not enough high quality evidence available to advocate the use of abdominal radiographs in children with symptoms of constipation.

Comment(s)

In order to use abdominal radiographs in the treatment of children with constipation there need to be a suitable method of reviewing radiographs. The different score that were investigated in the studies showed large intra-observer and large inter-observer variability. Maurice Van Den Bosch et al (2006) states that the Leech score had the lowest intra- and inter-observe variability and would be useful in the interpretation of radiographs. Whereas this is disputed by De Lorijin et al. Bongers M.E.J et al, states that the use of abdominal radiographs should be reserved for children in whom faecal impaction or constipation cannot be ruled out by other methods such as rectal examination. Reuchlin-Vroklage et al performed a high quality systematic analysis of the evidence that was available and came to the conclusion that the current research has yielded conflicting results. Reuklin-Vroklage et al came to the conclusion that Faecal loading on an abdominal radiograph had no diagnostic association with the clinical symptoms. The amount of radiation that a child receives when an abdominal radiograph is taken is not negligible and therefore abdominal radiographs should not routinely be performed in children who present to the Emergency department with constipation.

Clinical Bottom Line

Abdominal radiographs should not be routinely used in the evaluation of constipation in a child.

References

1. Maurice Van Den Bosch, et al, Systematic assessment of constipation on plain abdominal radiographs in children. Pediatric Radiology 36(3):224-6, 2006 Mar
2. Bongers M.E.J et al, The value of the abdominal radiograph in children with functional gastrointestinal disorders. European Journal of Radiology. 59(1):8-13, 2006 Jul.
3. Reuchlin-Vroklage L.M, et al, Diagnostic value of abdominal radiography in constipated children: a systematic review Archives of Pediatrics & Adolescent Medicine 159(7):671-8, 2005 Jul.