• 
• 

Three Part Question

In [young children presenting to the Emergency Department with abdominal pain] is [routine plain abdominal x-ray] useful at [formulating the final diagnosis]

Clinical Scenario

A 5 year old child comes into the Emergency Department with abdominal pain. On 'auto-pilot' you order a plain abdominal film. You wonder - will this actually help me make a diagnosis?

Search Strategy

MEDLINE 1950 – June week 1 2007 using the OVID interface
[{paediatric filter 2003 updated version for MEDLINE OVID interface} AND {exp Abdominal Pain/ OR exp Acute Abdomen/ OR abdo$ pain} AND {abdominal x-ray.mp OR exp Radiography, Abdominal/ OR abdo$ radio$} AND {exp Diagnosis, Differential/ OR exp Diagnosis/}] LIMIT to human AND English

Search Outcome

302 papers were found of which 297 were irrelevant or of insufficient quality.
The remaining 5 are shown in the table below:

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Zona JZ et al. 1975 USA	118 children (2-16) admitted with suspected appendicitis had abdominal and chest radiographs.	Cohort	% Appendicitis cases with radiographic features (appendicolithiasis, periappendiceal abscess)	93%	Limited data - no discussion of clinical suspicion, usefullness of radiograph, sensitivity or specificity. No gold standard. Type of study unclear. No independent reviewer of radiographs No Powers calculation
			% Pneumonia cases with radiographic features	100%
Rothrock SG et al. 1992 USA	354 children (<16 mean age 5.2) presenting to the Emergency Department who received abdominal radiography (regardless of presenting complaint).	Prospective observational study	Diagnostic radiographs (all cases vs major disease - surgical intervention necessary)	9% vs 38%	No data on single plain abdominal radiograph - a no. of views analysed No power calculation
			Suggestive features on radiograph	4% vs 10%
			Misleading features on radiograph	3% vs 2%
			Optimal indication for AXR - clinical suspician of major disease > 40%. Would detect most diagnostic XR in children with acute abdo disease if restricted to those with: prior abdo surgery, suspected FB ingestion, abn bowel sounds, abdo distension, or peritoneal signs.	sens 70% false +ves 39%
Bohner H et al. 1998 Germany	704 patients (age >6) presenting with acute abdominal pain, had a plain abdominal radiograph. In 79 of these patients bowel obstruction was suspected.	Prospective cohort study	Sensitivity in bowel obstruction	90.8%	Large proportion - 19.6% x-ray study results unknown Age range unspecified No subset analysis for children No Powers calculation
				ppv 80.2%
			For all cases of abdo pain: important findings on radiograph leading to diagnosis	15.8%
			Radiograph containing some useful information	10.4%
			No abnormal findings on radiograph	54.3%
Aloo J et al 2004 Canada	27 children under 3 years dagnosed with appendicitis - all underwent appendicectomy. 21 of the children had an Abdominal radiaograph.	Retrospective cohort	Useful signs found on radiograph	80.9%	Small study - no power calculation No values of sensitivity and specificity No discussion of severity of appendicitis Radiographs not independently reviewed
			(small bowel obstruction, fecalith, pneumoperitoneum)
Ulukaya Durakbaa C et al 2006 Turkey	424 children (11 months -17yr) with suspected appendicitis, all had plain erect abdo XR prior to surgery.	Retrospective cohort	Signs with high sensitivity but low specificity	calcified faecolith, mass image in RLQ, psoas obscuration, localised extraluminal air	No power calculation. No info on whether XR result altered diagnosis or management.
			signs most frequent in appendicits	dilated transverse colon (sens 53%, spec 72%), single air fluid level in RLQ (sens 48%, spec 76%)

Comment(s)

The papers agree that an abdominal film provides diagnostic information in over 10% of cases of abdominal pain. The overwhelming consensus is that they provide information supportive of the clinical diagnosis, however the vast majority do not alter the patient's diagnosis and management - and are therefore unnecessary. In cases where 'major disease' (requiring acute surgical intervention) is suspected (Rothrock SG et al) the diagnostic rate increases to 38%. A radiograph may then be indicated to confirm advanced disease where clinical suspicion is high. The radiograph is of use to detect small bowel obstruction, perforation, foreign bodies and calculi - with high specificity. Where these signs are present on a radiograph, a diagnosis can be confidently made. However due to low sensitivity the absence of signs does not rule out a diagnosis of obstruction. In cases of appendicitis there are a number of highly specific signs - calcified fecalith, psoas obscuration and localised extraluminal air (Ulukaya Durakbaa C et al) however incidence and sensitivity are low. The most common signs associated with appendicitis are normal radiograph, dilated transverse colon and single air fluid level in right lower quadrant. These signs have poor sensitivity and specificity - an abdominal radiograph for suspected appendicitis has low accuracy. When balancing radiation exposure with useful clinical information it must be concluded that the abdominal radiograph as a 'screening tool' for major disease is ineffective. It is of most use to confirm bowel obstruction where clinical suspicion is high, and to exclude things such as foreign body where there is a history of ingestion.

Clinical Bottom Line

A plain abdominal radiograph is not indicated as a screening tool for major disease in abdominal pain. It is best used where clinical suspicion of underlying major bowel disease (obstruction, perforation, foreign body etc) is high.

References

1. Zona JZ et al. Radiological aids in the diagnosis of appendicitis in children Southern Medical Journal 1975;68(11):1373-6
2. Rothrock SG et al. Plain abdominal radiography in the detection of major disease in children: a prospective analysis Annals of Emergency Medicine 1992;21(12):1423-9
3. Bohner H et al. Simple data from history and physical examination help to exclude bowel obstruction and to avoid radiographic studies in patients with acute abdominal pain European Journal of Surgery 1998;164(10):777-84
4. Aloo J et al Appendicitis in children less than 3 years of age: a 28-year review Pediatric Surgery International 2004;19(12):777-9
5. Ulukaya Durakbaa C et al An evaluation of individual plain abdominal radiography findings in pediatric apendicitis:results from a series of 424 children Turkish Journal of Trauma and Emergency Surgery 2006;12(1):51-8